Article
Computer Science, Artificial Intelligence
Hongzong Li, Jun Wang
Summary: This paper presents a collaborative annealing power k-means++ clustering algorithm, which integrates the k-means++ and power k-means algorithms in a collaborative neurodynamic optimization framework to select initial cluster centers and improve clustering performance.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Carlo Baldassi
Summary: We introduce an evolutionary algorithm called recombinator-k-means for optimizing the highly nonconvex kmeans problem. Its defining feature is that its crossover step involves all the members of the current generation, stochastically recombining them with a repurposed variant of the k-means++ seeding algorithm. The recombination also uses a reweighting mechanism that realizes a progressively sharper stochastic selection policy and ensures that the population eventually coalesces into a single solution. We compare this scheme with a state-of-the-art alternative, a more standard genetic algorithm with deterministic pairwise-nearest-neighbor crossover and an elitist selection policy, of which we also provide an augmented and efficient implementation. Extensive tests on large and challenging datasets (both synthetic and real word) show that for fixed population sizes recombinator-k-means is generally superior in terms of the optimization objective, at the cost of a more expensive crossover step. When adjusting the population sizes of the two algorithms to match their running times, we find that for short times the (augmented) pairwise-nearest-neighbor method is always superior, while at longer times recombinator-k-means will match it and, on the most difficult examples, take over. We conclude that the reweighted whole-population recombination is more costly but generally better at escaping local minima Moreover, it is algorithmically simpler and more general (it could be applied even to k-medians or k-medoids, for example).
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2022)
Article
Chemistry, Analytical
Ildeberto Santos-Ruiz, Francisco-Ronay Lopez-Estrada, Vicenc Puig, Guillermo Valencia-Palomo, Hector-Ricardo Hernandez
Summary: This paper presents a method for optimal pressure sensor placement in water distribution networks using information theory. The method selects sensor nodes based on maximizing information relevance and minimizing redundancy, resulting in improved leak detection. Experimental validation was conducted using pressure change data sets generated by simulation, and a heuristic algorithm was used for efficient sensor node selection.
Article
Engineering, Electrical & Electronic
Onat Gungor, Tajana S. Rosing, Baris Aksanli
Summary: This paper proposes a new approach for sensor placement in wireless sensor networks that takes into account the robustness of the system. Experimental results show that this method is more effective than existing approaches. The paper further improves the method by considering distance uncertainty, leading to an increased probability of target detection.
IEEE SENSORS JOURNAL
(2022)
Article
Environmental Sciences
Dina Zaman, Ashok Kumar Gupta, Venkatesh Uddameri, Manoj Kumar Tiwari, Dhrubajyoti Sen
Summary: A decision support framework based on network sensitivity analysis and multi-criteria decision-making was proposed to evaluate the prospect of leakage control through sensor placement in existing water distribution networks. The study focuses on leak control and optimization of sensor placement in water networks.
JOURNAL OF ENVIRONMENTAL MANAGEMENT
(2022)
Article
Computer Science, Artificial Intelligence
Hongzong Li, Jun Wang
Summary: This paper presents an upgraded version of the CAPKM++ clustering algorithm, called CAPKM++2.0. It introduces the re-initialization of weights and the iterative minimization of the power-mean sum during annealing. Ablation study results and experimental results on benchmark datasets demonstrate the effectiveness of the power-mean and the collaboration of multiple modules, and the superior clustering performance of the upgraded algorithm.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Chemistry, Multidisciplinary
Jan Y. K. Chan, Alex Po Leung, Yunbo Xie
Summary: The proposed method uses random projection to speed up kernel k-means clustering, significantly reducing computational complexity. Experiment results show a speed-up of 2 to 26 times with minimal performance degradation.
APPLIED SCIENCES-BASEL
(2021)
Article
Chemistry, Analytical
Qinghua Luo, Kexin Yang, Xiaozhen Yan, Jianfeng Li, Chenxu Wang, Zhiquan Zhou
Summary: This paper proposes a new trilateration algorithm based on combination and K-Means clustering, which effectively removes significant errors in the positioning results and utilizes the position and distance information of anchor nodes. Experimental results show that the proposed algorithm performs well in positioning accuracy and efficiency in different environments.
Article
Engineering, Environmental
Dinesh Kumar Gautam, Prakash Kotecha, Senthilmurugan Subbiah
Summary: A new method for optimizing sensor placement in water distribution networks using multi-objective optimization was proposed. Results from experiments on benchmark networks showed that the proposed algorithm outperformed the traditional method in terms of detection likelihood and Pareto points.
Article
Energy & Fuels
Kai Vahldiek, Bernd Rueger, Dennis Pierl, Frank Klawonn
Summary: The study delves into the optimal positioning and quantity of sensors in a district heating network. Through automation, an effective assessment and deployment of optimal sensor layout can be achieved.
Article
Computer Science, Artificial Intelligence
Yuan Ping, Huina Li, Bin Hao, Chun Guo, Baocang Wang
Summary: The study proposes a clustering method beyond k-means++, which achieves explicit cluster inference and utilization through local geometric information, and avoids the dependence on K through an iterative division and aggregation framework. Reasonability checking strategy and edge shrinkage strategy are designed to support arbitrary cluster shapes, improving clustering accuracy.
PATTERN RECOGNITION
(2024)
Article
Environmental Sciences
Tomer Shmaya, Avi Ostfeld
Summary: Water age refers to the time it takes for water to travel through a distribution system and reach the consumer. Higher water pressure results in shorter water age, while lower pressure leads to higher water age. This paper presents an algorithm for placing pressure reducing valves (PRVs) to reduce water age in water distribution systems.
Article
Environmental Sciences
Pham Duc Dai
Summary: Water loss due to water leakages in water distribution systems is a global challenge. Optimal pressure management through regulating operations of pressure reducing valves is crucial for water utilities.
Article
Chemistry, Multidisciplinary
Jie Meng, Ziyang Yu, Yuxin Cai, Xiuling Wang
Summary: Statistical methods were used to study the classification of high-potassium glass and lead-barium glass and analyze the correlation between the chemical composition of different glass samples. A categorization methodology for glass cultural relics was investigated, and a principal component analysis along with a case-specific clustering algorithm (K-Means++) was developed to further categorize the relics. The validity of the defined subcategories was verified, and the clustering algorithm's sensitivity was evaluated. The proposed algorithm and classification result provide a more comprehensive understanding of glass relics.
APPLIED SCIENCES-BASEL
(2023)
Article
Engineering, Chemical
Jose-Roberto Bermudez, Francisco-Ronay Lopez-Estrada, Gildas Besancon, Guillermo Valencia-Palomo, Ildeberto Santos-Ruiz
Summary: This work proposes a model predictive control strategy for pressure management and leakage reduction in a water distribution system. The proposed method considers the dynamic representation of the system and uses a pressure-reducing valve as a control element. It takes into account the nonlinearity of the valve, the demand profile, and pipeline leaks. The effectiveness of the proposed control system is demonstrated through various scenarios, achieving a reduction of approximately 5% in water losses and showing robustness in uncertain leak locations.
Article
Energy & Fuels
Erfan Mohagheghi, Mansour Alramlawi, Aouss Gabash, Frede Blaabjerg, Pu Li
Article
Engineering, Multidisciplinary
Mansour Alramlawi, Pu Li
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS
(2020)
Article
Engineering, Civil
Elad Salomons, Mashor Housh
JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT
(2020)
Article
Environmental Sciences
Elad Salomons, Lina Sela, Mashor Housh
WATER RESOURCES RESEARCH
(2020)
Article
Environmental Sciences
Mengning Qiu, Elad Salomons, Avi Ostfeld
Summary: This study proposes an analytical method for modeling the slug-feed method of disinfection and demonstrates its efficacy in two case studies, showing higher robustness compared to current standards and literature. The method provides additional resource utilization flexibility for water authorities and offers extra levels of district metered areas prioritization.
WATER RESOURCES RESEARCH
(2021)
Article
Environmental Sciences
Dionysios Nikolopoulos, Avi Ostfeld, Elad Salomons, Christos Makropoulos
Summary: Water distribution networks are critical infrastructure vulnerable to cyber-physical attacks, requiring tools and methodologies to assess system resilience and vulnerabilities. Deployment of water quality monitoring sensors can help protect against water contamination, but attacks on the monitoring system can lead to severe disruptions, emphasizing the need for risk management practices. The study demonstrates a novel methodology for assessing sensor placement schemes under simulated attacks, showing that some designs are more inherently resilient than others.
Article
Environmental Sciences
Elad Salomons, Uri Shamir, Mashor Housh
Summary: This study demonstrates a practical methodology to estimate pump curves using data collected by SCADA systems, allowing for continuous monitoring of pump performance in water distribution systems. In the era of IoT and big data, it is possible to monitor pump performance and derive updated pump characteristic curves without relying solely on periodic field tests.
Article
Engineering, Civil
Mohamad Zeidan, Pu Li, Avi Ostfeld
Summary: This study introduces a heuristic multiobjective approach for segmenting and operating water distribution systems. By balancing operational costs, excess pressure, and water age, multiobjective problems can be solved under a given system partitioning. However, enforcing clustering formation may compromise optimality and completeness, necessitating additional sensitivity analysis.
JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT
(2021)
Article
Ergonomics
Jan Schnee, Juergen Stegmaier, Pu Li
Summary: The study proposes an online approach based on IMU signals to classify bicycle crashes, functioning as a trigger for an automatic emergency system. Utilizing decision trees and probabilistic models, different kinematic events and crash scenarios are classified. The model's accuracy is verified through a series of driving tests, showing high sensitivity and specificity.
ACCIDENT ANALYSIS AND PREVENTION
(2021)
Article
Engineering, Civil
Mashor Housh, Elad Salomons, Lina Sela, Angus R. Simpson
Summary: Forum papers are opinion pieces or essays that are based on facts and sometimes contain speculation. They focus on civil engineering topics that are of general interest and relevance to the readership of the journal.
JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT
(2022)
Article
Engineering, Civil
Elad Salomons, Mashor Housh
Summary: Forum papers are opinion pieces or essays based on facts, sometimes containing speculation, that are relevant and interesting to the readership of the journal. The views expressed in this Forum article do not necessarily reflect the views of ASCE or the Editorial Board of the journal.
JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT
(2022)
Article
Environmental Sciences
Gopinathan R. Abhijith, Elad Salomons, Avi Ostfeld
Summary: Water distribution systems are vulnerable to various types of contamination, and online water-quality monitoring sensors are usually placed in selected locations as a preventive measure. However, the current state-of-the-art overlooks the impact of cyber-physical attacks on these sensor networks. This study aims to evaluate the effects of such attacks and provides valuable insights for decision makers in the water industry and regulators.
Article
Energy & Fuels
Gbalimene Richard Ileberi, Pu Li
Summary: Renewable energy resources and energy efficiency measures are effective ways to reduce CO2 emissions. This study investigates the potential of designing a hybrid system that incorporates hydrokinetic energy into an off-grid area. By applying a genetic algorithm to a small community in Nigeria, an optimal micro-grid configuration with PV panels, batteries, hydrokinetic turbines, converter, and a diesel generator is determined.
Article
Mathematics
Ruslan Voropai, Abebe Geletu, Pu Li
Summary: This work explores the use of stochastic model predictive control (MPC) for systems described by parabolic partial differential equations with random parameters. By solving the resulting high-dimensional chance constrained optimization problem using an inner-outer approximation method, the proposed MPC scheme automatically generates probability tubes, simplifying the derivation of feasible solutions.
Article
Environmental Sciences
Rahel Amare Kidanu, Maria Cunha, Elad Salomons, Avi Ostfeld
Summary: Water distribution network design is a complex multi-objective optimization problem. An improved version of NSGA II is proposed in this research, which utilizes three new-generation methods to target different regions of the Pareto front. The results show that the proposed algorithm outperforms the original NSGA II in terms of broadening the Pareto front solution range, increasing solution density, and discovering more non-dominated solutions.