An Improved Bio-Inspired Material Generation Algorithm for Engineering Optimization Problems Including PV Source Penetration in Distribution Systems

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(2), P. 603 - 603

Published: Jan. 9, 2025

The Material Generation Optimization (MGO) algorithm is an innovative approach inspired by material chemistry which emulates the processes of chemical compound formation and stabilization to thoroughly explore refine parameter space. By simulating bonding processes—such as ionic covalent bonds—MGO generates new solution candidates evaluates their stability, guiding toward convergence on optimal values. To improve its search efficiency, this paper introduces Enhanced (IMGO) algorithm, integrates a Quadratic Interpolated Learner Process (QILP). Unlike conventional random selection, QILP strategically selects three distinct compounds, resulting in increased diversity, more thorough exploration space, improved resistance local optima. adaptable non-linear adjustments QILP’s quadratic function allow traverse complex landscapes effectively. This IMGO, along with original MGO, developed support applications across phases, showcasing versatility enhanced optimization capabilities. Initially, both MGO algorithms are evaluated using several mathematical benchmarks from CEC 2017 test suite measure Following this, applied following well-known engineering problems: welded beam design, rolling element bearing pressure vessel design. simulation results then compared various established bio-inspired algorithms, including Artificial Ecosystem (AEO), Fitness–Distance-Balance AEO (FAEO), Chef-Based Algorithm (CBOA), Beluga Whale (BWOA), Arithmetic-Trigonometric (ATOA), Atomic Orbital Searching (AOSA). Moreover, IMGO tested real Egyptian power distribution system optimize placement PV capacitor units aim minimizing energy losses. Lastly, parameters estimation problem successfully solved via considering commercial RTC France cell. Comparative studies demonstrate that not only achieves significant loss reduction but also contributes environmental sustainability reducing emissions, overall effectiveness practical applications. outcomes 23 benchmark models average accuracy enhancement 65.22% consistency 69.57% method. Also, application achieved computational errors 27.8% while maintaining superior stability alternative methods.

Language: Английский

---

Generalizability of machine learning models for diabetes detection a study with nordic islet transplant and PIMA datasets

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Feb. 6, 2025

Diabetes Mellitus (DM) is a global health challenge, and accurate early detection critical for effective management. The study explores the potential of machine learning improved diabetes prediction using microarray gene expression data PIMA set. Researchers utilizing hybrid feature extraction method such as Artificial Bee Colony (ABC) Particle Swarm Optimization (PSO) followed by metaheuristic selection algorithms Harmonic Search (HS), Dragonfly Algorithm (DFA), Elephant Herding (EHA). Evaluated performance system following classifiers Non-Linear Regression—NLR, Linear Regression—LR, Gaussian Mixture Model—GMM, Expectation Maximization—EM, Bayesian Discriminant Analysis—BLDA, Softmax Classifier—SDC, Support Vector Machine with Radial Basis Function kernel—SVM-RBF classifier on two publicly available datasets namely Nordic Islet Transplant Program (NITP) Indian Dataset (PIDD). findings demonstrate significant improvement in classification accuracy compared to all genes. On islet transplant dataset, combined ABC-PSO EHO achieved highest 97.14%, surpassing 94.28% obtained ABC alone selection. Similarly, combination best 98.13%, exceeding 95.45% DFA These results highlight effectiveness our proposed approach identifying most informative features prediction. It observed that parametric values attained are almost similar. Therefore, this research indicates robustness FE FS along techniques different datasets.

Language: Английский

---

Solving Innovative Problems of Thrust Vector Control Based on Euler's Scientific Legacy

Civil Engineering Journal, Journal Year: 2023, Volume and Issue: 9(11), P. 2868 - 2895

Published: Nov. 1, 2023

This study aims to develop an interdisciplinary approach solving innovative thrust vector control problems. The methodology involves the development of a working hypothesis about ejection process when using controlled nozzle deflect (velocity vector) in any direction within complete geometric sphere. When developing hypothesis, multilateral analysis individual facts and scientific technical information is performed tools "big data" area, assessing opportunities apply "Foresight" for predicting fluidics. authors propose new mathematical models describe distribution mass flow rate fluid medium between channels. Patents inventions support novelty results that reveal more active fluidics as applied simple complex jet systems with low extremely high energy density flows. proposed rests on modern computer base logical continuation well-known Euler’s works. simulation multiflow devices mainly focuses power engineering, production, processing hydrocarbons. Some this research work, including patented design developments calculation methods, also robotics, unmanned vehicles, programable systems. attribute further inventive problems use different AI options. Doi: 10.28991/CEJ-2023-09-11-017 Full Text: PDF

Language: Английский

---

A Review of Path Planning Methods for Marine Autonomous Surface Vehicles

Journal of Marine Science and Engineering, Journal Year: 2024, Volume and Issue: 12(5), P. 833 - 833

Published: May 16, 2024

A marine autonomous surface vehicle (ASV) is a kind of robot with intelligent and flexible use advantages. They are mainly divided into two categories: unmanned vessels sailboats. Marine ASVs essential in science, industry, environmental protection, national defense. One the primary challenges faced by autonomously planning paths an intricate environment. Numerous research findings have surfaced recent years, including combination popular machine learning. However, systematic literature review still lacking, primarily comprehensive comparison types ASV path methods. This first introduces problem evaluation indicators for ASVs. Then, aiming at sailboats, respectively, it sorts out various algorithms proposed existing literature, advantages limitations both kinds ASVs, discusses them indicators. Also, this paper explores how factors affect its corresponding treatment Finally, summarizes ship planning, proposes potential technical solutions future development directions, aims to provide references further field.

Language: Английский

---

Advancements in optimizing wave energy converter geometry utilizing metaheuristic algorithms

Renewable and Sustainable Energy Reviews, Journal Year: 2024, Volume and Issue: 197, P. 114398 - 114398

Published: April 3, 2024

Language: Английский

---

Spark-based multi-verse optimizer as wrapper features selection algorithm for phishing attack challenge

Cluster Computing, Journal Year: 2024, Volume and Issue: 27(5), P. 5799 - 5814

Published: Feb. 12, 2024

Language: Английский

---

Integrating data-driven mechanisms for enhancing efficiency in business administration through biomechanics and bio-inspired modeling

Molecular & cellular biomechanics, Journal Year: 2025, Volume and Issue: 22(1), P. 614 - 614

Published: Jan. 10, 2025

This paper explores integrating biomechanics data and bio-inspired models to enhance efficiency in business administration, focusing on task scheduling, resource allocation, workflow optimization. Biomechanics, traditionally applied fields such as healthcare sports, is used analyze human movement physical strain processes, particularly physically demanding environments like manufacturing logistics. Bio-inspired models, Genetic Algorithms (GA) Particle Swarm Optimization (PSO), are solve complex optimization problems management scheduling. The study presents three case studies demonstrate the practical application of these methodologies: (1) a environment using reduce improve completion times; (2) allocation Supply Chain Management (SCM) PSO minimize transportation labor costs while improving warehouse utilization delivery (3) scheduling an office GA efficiency, workload distribution, employee satisfaction. results 21.6% reduction shoulder joint 18.2% improvement time setting; 16.1% 18.6% SCM PSO; 17.6% decrease makespan 29.8% distribution through GA-based environment. These findings underscore potential combining human-centered with operational well-being, cost-effectiveness significantly.

Language: Английский

---

Generalized Gene Selection for Microarray Classification Via Improved Crested Porcupine Optimizer

Published: Jan. 1, 2025

Language: Английский

---

Matching Community Sports Facilities With Ant Colony Algorithm in National Fitness

International Journal of Distributed Systems and Technologies, Journal Year: 2025, Volume and Issue: 16(1), P. 1 - 20

Published: Feb. 15, 2025

This study addresses the challenge of selecting optimal locations for urban sports facilities, leveraging strengths ant colony optimization (ACO) algorithm. An enhanced ACO model is proposed, incorporating population density and distance to facilities as critical factors in objective function. The employs a unique pheromone updating strategy that reduces search time improves solution quality. Two updates levels are performed, initial distribution reset based on path distances. effectiveness demonstrated through case Yuhua District, Changsha City, where it successfully identifies prime public facilities. research contributes literature facility siting planning by offering practical optimizing infrastructure within cities.

Language: Английский

---

From collective intelligence to global optimisation: an agent-based model approach

Computing, Journal Year: 2025, Volume and Issue: 107(3)

Published: Feb. 27, 2025

Language: Английский

---