Optimal Clustering Research Articles

A reconfigurable architecture for maximizing energy harvesting of thermoelectric generators in non-stationary conditions

The use of Thermoelectric Generators (TEGs) has proliferated across a multitude of applications for energy harvesting. As more modules are employed to recover greater amounts of energy, the temperature mismatch between them increases. This results in each module operating at a distinct maximum power point, thereby reducing the overall system efficiency. Furthermore, in dynamic applications such as automotive scenarios, the temperatures of the thermoelectric generators are not constant, and the maximum power point accordingly shifts. A fixed architecture is unable to cope with these fluctuating situations. Therefore, this paper introduces a reconfigurable architecture capable of harnessing maximum energy at any given moment, improving energy recovery compared to a fixed architecture. Optimization techniques, lean methodologies, and clustering approaches are employed to efficiently design the reconfigurable TEG, which enables modification of the electrical connections inside the TEG modules and the number of Maximum Power Point Tracking (MPPT) modules. A use case is presented where the reconfigurable TEG is compared with fixed, yet optimized, TEG configurations under mixed driving modes. In this specific case, the results demonstrate that the reconfigurable TEG achieves enhanced performance in dynamic environments with two MPPTs under mixed scenarios, reaching an efficiency of 96.3% and a 0.29% improvement in energy recovery.

MULTILEVEL MODAL VALUE ANALYSIS FOR INTERPRETING CATEGORICAL K-MEDOIDS CLUSTERS DATA

Consumer segmentation plays a crucial role for business owners in developing their enterprises. K-Medoid is commonly used for segmentation functions due to its low computational complexity. However, K-Medoid has limitations, such as the variability in cluster sizes across different iterations and the challenge of determining the optimal number of clusters. The Davies-Bouldin Index (DBI) is a metric used to evaluate the number of clusters by calculating the ratio between the within-cluster distance and the between-cluster distance. Most segmentation studies typically stop at the formation of clusters without further interpretation, particularly when dealing with categorical data. This study aims to modify the use of K-Medoid and propose a method for interpreting clusters with categorical data. The research began with questionnaire design and the data collecting from 100 respondents, which was normalized in the second stage. Clustering used K-Medoid with variations K values from K=2 to K=10, with each K value tested 10 times. The clustering results were evaluated using the DBI to select the optimal clusters. Data interpretation conducted using modal values, calculated as the ratio of the number of times a specific attribute variable was selected by respondents to the total number of data points in the cluster. Utilization and hierarchical visualization of modal values proposed in this study offer insights into the dominant variables within an attribute and also depict the relationships between attributes based on the ranking of modal values. These advantages facilitate business analysts in labeling clusters for developing consumer-driven business strategies.

Achieving the Best Symmetry by Finding the Optimal Clustering Filters for Specific Lighting Conditions

This article explores the efficiency of various clustering methods for image segmentation under different luminosity conditions. Image segmentation plays a crucial role in computer vision applications, and clustering algorithms are commonly used for this purpose. The search for an adaptive clustering mechanism aims to ensure the maximum symmetry of real objects with objects/segments in their digital representations. However, clustering method performances can fluctuate with varying lighting conditions during image capture. Therefore, we assess the performance of several clustering algorithms—including K-Means, K-Medoids, Fuzzy C-Means, Possibilistic C-Means, Gustafson–Kessel, Entropy-based Fuzzy, Ridler–Calvard, Kohonen Self-Organizing Maps and MeanShift—across images captured under different illumination conditions. Additionally, we develop an adaptive image segmentation system utilizing empirical data. Conducted experiments highlight varied performances among clustering methods under different luminosity conditions. This research enhances a better understanding of luminosity’s impact on image segmentation and aids the method selection for diverse lighting scenarios.

Delay sensitive and energy adaptive clustering hierarchy protocol using enhanced agglomerative hierarchy algorithm with time‐controlled jellyfish optimization

AbstractWireless sensor networks have sensing functions which made up of small sensor nodes that are processing and communication capabilities. However, the traditional approach of designating a single sensor node as the cluster head often leads to a faster depletion of its energy resources than another node. So, periodic reassignment of the cluster head role among different sensor nodes is crucial to extend operational lifetime and ensure sustained performance. Therefore, this research proposes an Enhanced Agglomerative Hierarchy Algorithm based on the Low Energy Adaptive Clustering Hierarchy (EAHALEACH) protocol, coupled with an enhanced time‐controlled optimization algorithm, to enhance energy efficiency by selecting the optimal cluster head from a candidate pool. The proposed method enhances energy efficiency by optimally selecting cluster heads, which reduces energy consumption and extends the lifespan of the network. Additionally, a lightweight energy‐aware cluster head rotation algorithm is introduced to efficiently rotate the cluster head role within the sensor node network, minimizing unnecessary data transmissions and extending the network lifetime. A hybrid approach combining Carrier Sense and Time Division Multiple Access (CSTDMA) optimizes packet forwarding by dynamically allocating time slots, reducing collisions and contention to improve packet forwarding efficiency. Comparative analysis with existing techniques demonstrates that the hybrid clustering‐based LEACH protocol achieves superior performance in terms of throughput 4600 bps by 2000 rounds, energy consumption 50 J, latency as 1 ms and network lifetime in 1000th node attains 3500 s. These advancements contribute to prolonged operational efficiency and sustained performance in wireless sensor network deployments.

A Novel Approach for Optimal Cluster Identification and N-Order Hesitation Based Time Series Forecasting

A Novel Approach for Optimal Cluster Identification and N-Order Hesitation Based Time Series Forecasting

Bayesian optimization of proton generation in terawatt laser–CH2 cluster interactions within a plasma channel

Improving the energy efficiency in generating high-energy proton or boron ions is crucial for advancing the feasibility of neutronless laser-based proton–boron (p-B11) fusion reactions. The primary objective of this work is to optimize the fusion energy efficiency of a proposed advanced p-B11 fusion scheme. In the proposed scheme, an ultrashort laser pulse is guided by a plasma channel filled with carbon–hydrogen (CH2) clusters. The MeV protons are generated by the Coulomb explosion (CE) of the cluster, which, therefore, interact with surrounding boron to produce alpha particles. To evaluate the fusion energy efficiency under various conditions, 2D particle-in-cell (PIC) simulations are used, supplemented with analytical calculations and estimations. The Bayesian optimization (BO) algorithm is utilized to optimize the key interaction parameters. The BO approach allows us to identify optimal cluster and laser parameters that would have higher fusion energy efficiency.

Neuro‐fuzzy‐based cluster formation scheme for energy‐efficient data routing in IOT‐enabled WSN

SummaryInternet of things–enabled wireless sensor networks face challenges like inflexibility, poor scalability, suboptimal cluster head selection, and energy inefficiencies. This is due to the faster data transmission rates between cluster nodes during data packet routing. This creates unnecessary energy consumption burdens for those actively transmitting nodes. Conceptually, an effective cluster formation phase supports better data routing mechanisms, while sustaining the energy efficiency of individual nodes. This paper proposes a Neuro‐Fuzzy based Cluster Formation (NFCF) scheme to facilitate adaptive and energy‐efficient cluster topologies. NFCF utilizes fuzzy logic and neural networks to identify optimal super nodes for flexible cluster formations. This approach enables configurable cluster sizes along with inclusion/exclusion criteria for member nodes based on energy thresholds. Parameters evaluated for node selection include the degree of super node, expected energy per cluster, energy variance, and residual energy. Nodes not meeting the thresholds are excluded. The neural network updates fuzzy rules to guide optimal clustering decisions based on anticipated energy dynamics under different conditions. The performance of the proposed NFCF scheme is evaluated based on objective function changes related to data transmission, individual node energy variation, energy variance before and after transmissions, and averaged end‐to‐end delay across transmission cycles. Results are compared against genetic fuzzy clustering, fuzzy energy‐aware clustering, fuzzy‐based distributed clustering, fuzzy logic‐based multi‐hop clustering, and fuzzy weighted k‐means clustering.

Enhanced Parameter Estimation of DENsity CLUstEring (DENCLUE) Using Differential Evolution

The task of finding natural groupings within a dataset exploiting proximity of samples is known as clustering, an unsupervised learning approach. Density-based clustering algorithms, which identify arbitrarily shaped clusters using spatial dimensions and neighbourhood aspects, are sensitive to the selection of parameters. For instance, DENsity CLUstEring (DENCLUE)—a density-based clustering algorithm—requires a trial-and-error approach to find suitable parameters for optimal clusters. Earlier attempts to automate the parameter estimation of DENCLUE have been highly dependent either on the choice of prior data distribution (which could vary across datasets) or by fixing one parameter (which might not be optimal) and learning other parameters. This article addresses this challenge by learning the parameters of DENCLUE through the differential evolution optimisation technique without prior data distribution assumptions. Experimental evaluation of the proposed approach demonstrated consistent performance across datasets (synthetic and real datasets) containing clusters of arbitrary shapes. The clustering performance was evaluated using clustering validation metrics (e.g., Silhouette Score, Davies–Bouldin Index and Adjusted Rand Index) as well as qualitative visual analysis when compared with other density-based clustering algorithms, such as DPC, which is based on weighted local density sequences and nearest neighbour assignments (DPCSA) and Variable KDE-based DENCLUE (VDENCLUE).

Threshold-driven K-means sector clustering algorithm for wireless sensor networks

The clustering algorithm is an effective method for developing energy efficiency routing protocol for wireless sensor networks (WSNs). In clustered WSNs, cluster heads must handle high traffic, thus consuming more energy. Therefore, forming balanced clusters and selecting optimal cluster heads are significant challenges. The paper proposes a sector clustering algorithm based on K-means called KMSC. KMSC improves efficiency and balances the cluster size by employing symmetric dividing sectors in conjunction with K-means. For the selection of cluster heads (CHs), KMSC uses the residual energy and distance to calculate the weight of the node, then selects the node with the highest weight as CH. A hybrid single-hop and multi-hop communication is utilized to reduce long-distance transmissions. Furthermore, the impact of the number of sectors, the threshold for clustering, and the network size on the performance of KMSC has been explored. The simulation results show that KMSC outperforms EECPK-means, K-means, TSC, LSC, and SEECP in terms of FND, HND, and LND.

Robust and Scalable Federated Learning Framework for Client Data Heterogeneity Based on Optimal Clustering

Federated learning is a promising paradigm for applications across a variety of domains. However, there are some challenges that must be addressed in real-world scenarios, particularly the data heterogeneity among participating clients. Most existing studies primarily focus on the issue of non-independent and identically distributed data, but they do not consider the critical aspect of data quality heterogeneity. When low-quality data is contributed by some clients, the efficacy of models trained through the traditional approaches will be significantly compromised. Therefore, we propose ROSCFL, a robust and scalable federated learning framework for client data heterogeneity based on optimal clustering. We first develop a cluster contribution evaluation strategy based on the optimal clustering to quantify the contribution of each cluster. Next, we design a robust model aggregation strategy, which effectively mitigates the impact of low-quality data on the global model by optimizing weight allocation and client sampling. Finally, we introduce a client incorporation mechanism to enhance the scalability of ROSCFL. Extensive experiments have been conducted, and the results demonstrate that ROSCFL achieves strong robustness and scalability, particularly in scenarios wherein data distribution and quality heterogeneity coexist.

Fast Optimization of the Net Present Value of Unconventional Wells Using Rapid Rate-Transient Analysis

Summary Decline-curve analysis (DCA) is the industry standard to predict hydrocarbon production to then estimate the net present value (NPV) of unconventional wells. Conventional DCA assumes the flowing bottomhole pressure (BHP) is constant. This is an unrealistic assumption for many unconventional wells that can lead to incorrect estimates of ultimate recovery and thus erroneous NPV calculations. This work illustrates the application of a novel technique that combines variable BHP conditions with decline-curve models [Rapid rate-transient analysis (RTA)] to estimate the NPV and select the optimal cluster spacing and number of fractures for a given well. We compare the results of DCA and Rapid RTA to estimate and optimize the cluster spacing for a tight oil and shale gas well. The Rapid RTA results show marked differences in terms of the shape and the optimal value of the NPV function from the simpler DCA method. For the two cases illustrated, DCA results are a direct consequence of the rate-time analysis failing to correctly detect the onset of boundary-dominated flow (BDF). In contrast, Rapid RTA correctly detects the onset of BDF and presents a defined maximum value of the NPV vs. cluster spacing (number of fractures) function. This optimal value is a trade-off between the fracture treatment cost, the speed of recovery of hydrocarbons, and the value of money with time (discount rate). The major contribution of this work is the implementation of the Rapid RTA technique allowing fast computation of pressure-rate-time analysis and NPV calculations with computational times comparable with DCA for optimizing fracture cluster spacing and the total number of fractures of unconventional wells. We have developed a web-based application to give readers a hands-on experience of this new technique and to analyze and optimize the NPV of unconventional wells.

Customer Segmentation Using the K-means Clustering Algorithm and Recency Frequency Monetary Model at Pharmaceutical Product Wholesaler

PT Kimia Farma Trading and Distribution (KFTD) is a company engaged in the distribution and trading services of Indonesian health products, on a national scale. In 2022, the company aims to increase sales to be awarded as one of the top 3 national pharmaceutical product distributors by 2024. Their current strategy is to provide customers with delayed payment permission and integrated complaint services. However, the offers and services are the same for all customers which does not consider customer track record hence it is not cost-effective. One way to increase sales is by enhancing customer satisfaction and loyalty by implementing Customer Relationship Management (CRM) strategies. Several strategies can be carried out, namely analysis of associations related to pharmaceutical products, and analysis of customer segmentation and clustering of products. The method used in this study was the K-means clustering algorithm combined with the Recency Frequency Monetary (RFM) model. Experiments showed that the optimal clustering results are 4 therefore they are categorized into 4 customer segments, namely Superstar, Golden, Typical, and Occasional Customers.

Optimal cluster head localization for cluster-based wireless sensor network using free-space optical technology and genetic algorithm optimization

Optimal cluster head localization for cluster-based wireless sensor network using free-space optical technology and genetic algorithm optimization

OCAE-based feature extraction and cluster analysis of high-energy-consuming plant loads

Currently, in power systems and energy management, the extraction and clustering of load features play a critical role. It can support the planning, scheduling and equipment design of power systems, holding significant theoretical importance and practical benefits. Addressing the complexity of high-dimensional daily load data poses challenges in effectively capturing the intricate features of load sequence data and achieving optimal clustering outcomes. This paper proposes an optimal convolutional autoencoder (OCAE) model designed for good feature extraction. First, a large number of test comparisons demonstrate that the OCAE effectively mines deep feature data from load sequences and exhibits strong data reconstruction capabilities. Second, the OCAE model is compared with other autoencoders to validate its superior feature extraction performance. Finally, the deep feature information extracted by the OCAE encoder is utilized in the k-means clustering algorithm for cluster analysis and is quantitatively evaluated against other methods using Davies–Bouldin index (DBI) and silhouette coefficient (SC). The experimental results confirm the superiority of OCAE joint k-means clustering over improved 1D-CAE joint k-means clustering, with a reduction of approximately 12.3 % in DBI and an improvement of approximately 11.1 % in SC. The outstanding clustering outcomes achieved through OCAE joint k-means clustering have significant scientific implications for power system dispatch planning.

Application of the Agilent 2100 Bioanalyzer instrument as quality control for next-generation sequencing.

Next-generation sequencing (NGS) technologies have expanded the spectrum of forensic DNA analysis by facilitating efficient and precise genotyping of a large number of genetic markers. Yet, challenges persist regarding complex sample processing and assurance of equal molar concentrations across pooled samples. Since optimal cluster density is crucial for sequencing performance, the determination of both quantity and quality is indispensable for library preparation. In this study, we investigated the application of the Agilent 2100 Bioanalyzer for library quality control, as studies for forensic approaches, particularly for highly degraded postmortem samples, are rare. Our analysis encompassed assessing total DNA concentrations, fluorescence unit (FU) values, and adapter dimer concentrations in purified DNA libraries derived from buccal swabs and tissue samples of decomposed corpses. The sensitivity study tested a serial dilution derived from buccal swabs and revealed a decrease in FU values and an increase in adapter dimers with declining DNA input concentrations. Deviations in total DNA concentrations and average peak heights between the Agilent 2100 Bioanalyzer runs indicated a lack of repeatability in data and presented challenges in accurate quantification, which was also observed in previous studies. Yet, the analysis of degraded samples from decomposed human remains has shown the ability to detect adapter dimer concentrations, which can be crucial for the quality of subsequent NGS library preparation and sequencing success. Therefore, the Agilent 2100 Bioanalyzer proves to be a valuable tool for NGS quality control.

Optimized Adaptive Multi-Scale Dual An for Multi-Objective CHS and Energy-Aware Routing in 6G WC

In this research work, an Adaptive Multi-Scale Dual Attention Network with ZOA for Multi-Objective CHS with energy-aware routing in 6G wireless Communication (CHS-EAR-AM-SDAN-6G) is proposed to secure the data transmission by selecting optimum cluster heads in the 6G Wireless Communication network. Initially, the nodes are gathered together to form a cluster using an Adaptive Multi-Scale Dual Attention Network (AM-SDAN). The Zebra Optimization Algorithm (ZOA) strategically selects Cluster Heads (CHs) in wireless networks based on a multi-objective fitness function (MoFF) that minimizes energy consumption while considering factors, like distance, delay, and traffic density. The path and minimum value of fitness are recognized as the routing path and statistics are promoted to the sink node through the cluster head. The proposed scheme has been applied in Python and productivity of the proposed method is predictable with the help of several performances they are energy consumption, detection rate, computational time, packet delivery rate, number of alive nodes, and security. The performance of the proposed CHS-EAR-AMSDAN-6G method attains 25.93%, 24.81%, and 23.38% of alive nodes, 24.45%, 26.71% and 21.32% lower packet delivery rate, 27.56%, 26.43%, and 28.61% low computational period, related with three current methods, such as Generative Adversarial Learning for ITM in 6G Wireless Communication Networks (CHS-EAR-GAL-ITM-6G), ML Algorithms for the Future 6G WCN (CHS-EAR-ML-6G), Energy Efficient Distributed Federated Learning to the 6G Wireless Communication Networks (CHS-EAR-EEDFL-6G), respectively.

K-Means text clustering method based on Decision Grey Wolf Optimization

Aiming at the problem that the traditional algorithm is easy to fall into local optimum in the process of text clustering, which leads to inaccurate text clustering results, a text clustering method based on decision grey wolf optimization K-Means is proposed to cluster the text data set and the standard UCI data set respectively. Afterword segmentation, stop words removal, feature extraction, and text vectorization of text data, the powerful optimization ability of the Decision Gray Wolf Optimization (DGWO) algorithm is used for global optimization, and the clustering center in K-Means algorithm is replaced by the location of wolves. The position of the wolf group is updated by iterative optimization to obtain the optimal clustering center, to perform text clustering. The experimental results show that compared with the traditional method, the precision, recall, and F-Measure of the text data clustering are improved by 49.22%, 51.15%, and 48.98% respectively. The precision, recall, and F-Measure of UCI data clustering are increased by 23.92%, 25.40%, and 24.70% respectively, and the text clustering results are more reliable.

Co-regularized optimal high-order graph embedding for multi-view clustering

Real-world applications frequently involve multiple data modalities in the same samples, which are regarded as multi-view data. Multi-view clustering has been studied extensively in recent years to demonstrate embedded heterogeneity. However, most existing methods emphasize low-order correlation in multiple views, whereas approaches that incorporate high-order correlation are limited by the equal view-specific significance problem or a trade-off between global and local consistency. In this paper, we propose a co-regularized optimal graph-based clustering method known as Co-MSE, which integrates the correlation of different orders. By integrating the first-order and second-order similarities, the local structure is preserved, while an optimized embedding representation for multi-view data is obtained simultaneously through co-regularization. We demonstrate that Co-MSE can aid in providing a more suitable embedding representation and further enable satisfactory clustering performance. Extensive experiments on real-world datasets confirm the effectiveness and advantages of the proposed method.

Identification of generator coherency in power systems with wind farm

In the context of the "dual carbon" goal, the penetration rate of new energy represented by wind power is gradually increasing. The large-scale grid connection of wind power makes the power system more complex and uncertain. The output of wind farms changes the system flow, indirectly affecting the power angle characteristics of synchronous generators, and thereby changing the coherence between generators. Affects synchronous generator homology identification. This paper proposes a generator homology identification method for power systems containing wind farms, in response to the problem that the existing methods for identifying unit homology have not taken into account the adverse effects of wind farms on identification results. Translate the influence of wind farms on the homology of synchronous generators into the contraction admittance matrix as a static electrical distance indicator; Select dynamic data reflecting the homology of synchronous generators after disturbance, and form four dynamic indicators to measure the power angle increment curve between each generator: Euclidean distance, Chebyshev distance, grey correlation, and correlation coefficient; By using the combination weighting method to determine the weights of each indicator, a comprehensive similarity matrix is formed, and the optimal clustering results are determined using fuzzy system clustering and F-statistical values. Finally, the effectiveness of the proposed method was verified using EPRI-9 node system, EPRI-36 node system, and IEEE68 node system as examples.© 2017 Elsevier Inc. All rights reserved.

Multi-Objective Unsupervised Feature Selection and Cluster Based on Symbiotic Organism Search

Unsupervised learning is a type of machine learning that learns from data without human supervision. Unsupervised feature selection (UFS) is crucial in data analytics, which plays a vital role in enhancing the quality of results and reducing computational complexity in huge feature spaces. The UFS problem has been addressed in several research efforts. Recent studies have witnessed a surge in innovative techniques like nature-inspired algorithms for clustering and UFS problems. However, very few studies consider the UFS problem as a multi-objective problem to find the optimal trade-off between the number of selected features and model accuracy. This paper proposes a multi-objective symbiotic organism search algorithm for unsupervised feature selection (SOSUFS) and a symbiotic organism search-based clustering (SOSC) algorithm to generate the optimal feature subset for more accurate clustering. The efficiency and robustness of the proposed algorithm are investigated on benchmark datasets. The SOSUFS method, combined with SOSC, demonstrated the highest f-measure, whereas the KHCluster method resulted in the lowest f-measure. SOSFS effectively reduced the number of features by more than half. The proposed symbiotic organisms search-based optimal unsupervised feature-selection (SOSUFS) method, along with search-based optimal clustering (SOSC), was identified as the top-performing clustering approach. Following this, the SOSUFS method demonstrated strong performance. In summary, this empirical study indicates that the proposed algorithm significantly surpasses state-of-the-art algorithms in both efficiency and effectiveness. Unsupervised learning in artificial intelligence involves machine-learning techniques that learn from data without human supervision. Unlike supervised learning, unsupervised machine-learning models work with unlabeled data to uncover patterns and insights independently, without explicit guidance or instruction.