Optimization-based Algorithm Research Articles

Pressure Drop Estimation of Two-Phase Adiabatic Flows in Smooth Tubes: Development of Machine Learning-Based Pipelines

The current study begins with an experimental investigation focused on measuring the pressure drop of a water–air mixture under different flow conditions in a setup consisting of horizontal smooth tubes. Machine learning (ML)-based pipelines are then implemented to provide estimations of the pressure drop values employing obtained dimensionless features. Subsequently, a feature selection methodology is employed to identify the key features, facilitating the interpretation of the underlying physical phenomena and enhancing model accuracy. In the next step, utilizing a genetic algorithm-based optimization approach, the preeminent machine learning algorithm, along with its associated optimal tuning parameters, is determined. Ultimately, the results of the optimal pipeline provide a Mean Absolute Percentage Error (MAPE) of 5.99% on the validation set and 7.03% on the test. As the employed dataset and the obtained optimal models will be opened to public access, the present approach provides superior reproducibility and user-friendliness in contrast to existing physical models reported in the literature, while achieving significantly higher accuracy.

A CeO2 (100) surface reconstruction unveiled by in situ STEM and particle swarm optimization techniques.

The reconstruction of the polar CeO2 (100) surface has been a subject of long-standing debates due to its complexity and the limited availability of experimental data. Herein, we successfully reveal a CeO2 (100)-(4 × 6) surface reconstruction by combining in situ spherical aberration-corrected scanning transmission electron microscopy, density functional theory calculations, and a particle swarm optimization-based algorithm for structure searching. We have further elucidated the stabilizing mechanism of the reconstructed structure, which involves the splitting of the filled Ce(4f) states and the mixing of the lower-lying ones with the O(2p) orbitals, as evidenced by the projected density of states. We also reveal that the surface chemisorption properties toward water molecules, an important step in numerous heterogeneous catalytic reactions, are enhanced. These insights into the distinct properties of ceria surface pave the way for performance improvements of ceria in a wide range of applications.

ICG signal denoising based on ICEEMDAN and PSO-VMD methods.

Impedance cardiography (ICG) plays a crucial role in clinically evaluating cardiac systolic and diastolic functions, along with various other cardiac parameters. However, its accuracy heavily depends on precisely identifying feature points reflecting cardiac function. Moreover, traditional signal processing techniques used to mitigate random noise and breathing artifacts may inadvertently distort the amplitude and temporal characteristics of ICG signals. To address this issue, this study investigates a noise and artifact elimination method based on Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) and Particle Swarm Optimization-based Variational Mode Decomposition Algorithm (PSO-VMD). The goal is to preserve the amplitude and temporal features of ICG signals to ensure accurate feature point extraction and computation of associated cardiac parameters. Comparative analysis with signal processing methods employing various wavelet families and Ensemble Empirical Mode Decomposition (EEMD) in ICG signal processing applications reveals that the proposed method achieves superior signal-to-noise ratio (SNR) and lower root-mean-square error (RMSE), while demonstrating enhanced correlation and waveform consistency with the original signal.

Fault-Tolerant Optimal Consensus Control for Heterogeneous Multi-Agent Systems

This study explores fault-tolerant consensus in leader–following heterogeneous multi-agent systems, focusing on actuator failures in uncrewed aerial vehicles (UAVs) and uncrewed ground vehicles (UGVs). An optimization-based fault-tolerant consensus algorithm is proposed. The algorithm utilizes the Euler–Lagrange formula to ensure system consistency under actuator failures, with the Lyapunov stability theory proving the asymptotic stability of the consistency error. The algorithm is applied to heterogeneous multi-agent systems of UAVs and UGVs to derive optimal fault-tolerant consensus control laws for each vehicle type. Simulation experiments give evidence for the feasibility of the proposed control strategy.

Genetic algorithm-based optimization of combined supercritical CO2 power and flash-tank enhanced transcritical CO2 refrigeration cycle for shipboard waste heat recuperation

The shipping industry significantly contributes to global trade and economy, but is also responsible for approximately ∼3 % of global greenhouse gas (GHG) emissions. Harnessing waste heat from marine vessels to perform useful work is a potential solution to reduce these emissions. Although the Organic Rankine Cycle (ORC) and CO2 cycles have been studied separately for waste heat recovery, a combined power and refrigeration systems, utilizing both cycles have not been reported. Herein, a combined system, “Marine Energy Recovery System (MERS)”, that integrates the supercritical CO2 Brayton power cycle with a flash-tank-enhanced transcritical CO2 refrigeration cycle, is proposed, aiming to generate power and provide cooling simultaneously. The power cycle incorporates an ORC, and a flash-tank is introduced within the refrigeration cycle to improve system performance. Both cycles share a low-temperature recuperator and a gas cooler to further utilize the heat between them. The MERS is evaluated from both a 1st and 2nd law perspective against various performance metrics under different operating conditions, such as: gas cooler pressure, evaporation temperature, and turbine inlet temperature and pressure. Results elucidated that the MERS achieved energy and exergy efficiencies of 54.62 % and 54.90 %, respectively, representing significant improvements over similar systems. The net work output improved by 717.55 kW and the net cooling capacity by 515.7 kW relative to the reference system. Furthermore, a COP of 2.99 outlines its potential as a cooling system. To further enhance MERS’s performance, a multi-objective optimization approach is employed, utilizing Artificial Neural Network (ANN) and Genetic Algorithm (GA). Optimal operational conditions yielded an energy efficiency of 74.95 % while maintaining a net work output of 6890.55 kW with gas cooler pressure, turbine inlet temperature, and gas cooler outlet temperature being 9.5 MPa, 461.76°C and 50°C respectively. These findings support the reliability and improved design of the MERS for future marine applications.

A Survey on Information Bottleneck.

This survey is for the remembrance of one of the creators of the information bottleneck theory, Prof. Naftali Tishby, passing away at the age of 68 on August, 2021. Information bottleneck (IB), a novel information theoretic approach for pattern analysis and representation learning, has gained widespread popularity since its birth in 1999. It provides an elegant balance between data compression and information preservation, and improves its prediction or representation ability accordingly. This survey summarizes both the theoretical progress and practical applications on IB over the past 20-plus years, where its basic theory, optimization, extensive models and task-oriented algorithms are systematically explored. Existing IB methods are roughly divided into two parts: traditional and deep IB, where the former contains the IBs optimized by traditional machine learning analysis techniques without involving any neural networks, and the latter includes the IBs involving the interpretation, optimization and improvement of deep neural works (DNNs). Specifically, based on the technique taxonomy, traditional IBs are further classified into three categories: Basic, Informative and Propagating IB; While the deep IBs, based on the taxonomy of problem settings, contain Debate: Understanding DNNs with IB, Optimizing DNNs Using IB, and DNN-based IB methods. Furthermore, some potential issues deserving future research are discussed. This survey attempts to draw a more complete picture of IB, from which the subsequent studies can benefit.

Text document clustering using mayfly optimization algorithm with k-means technique

Text clustering is a subfield of machine learning (ML) and natural language processing (NLP) that consists of grouping similar sentences or documents based on their content. However, insignificant features in the documents minimize the accuracy of information retrieval which makes it challenging for the clustering approach to efficiently cluster similar documents. In this research, the mayfly optimization algorithm (MOA) with a k-means approach is proposed for text document clustering (TDC) to effectively cluster similar documents. Initially, the data is obtained from Reuters-21678, 20-Newsgroup, and BBC sports datasets, and then pre-processing is established by stemming and stop word removal to remove unwanted phrases or words. The data imbalance approach is established using an adaptive synthetic sampling algorithm (ADASYN), then term frequency-inverse document frequency (TD-IDF) and WordNet features are employed for extracting features. Finally, MOA with the K-means technique is utilized for TDC. The proposed approach achieves better accuracy of 99.75%, 99.54%, and 98.24% when compared to the existing techniques like fuzzy rough set-based robust nearest neighbor-convolutional neural network (FRS-RNN-CNN), TopicStriker, Modsup-based frequent itemset, and rider optimization-based moth search algorithm (Modsup-Rn-MSA), hierarchical dirichlet-multinomial mixture, and multi-view clustering via consistent and specific non-negative matrix (MCCS).

Optimization of magnetic circuit in electro-controlled permanent magnet blank holder process with magnetorheological elastomers and analysis of its impact on deep drawing

Addressing the issue of blank holder force (BHF) attenuation caused by the blank-holder gap in the EPMBH process, this study proposes a solution: employing magnetorheological elastomer (MRE) as a magnetic medium to fill these gaps, thus constructing an optimized edging magnetic circuit aimed at reducing magnetic loss and enhancing BHF. Initially, the study utilizes finite element analysis to build a dataset, followed by the application of a particle swarm to optimization BP neural network for magnetic force prediction and modeling. This process reveals the coupling relationship between the working gap thickness, MRE's relative permeability, and magnetic force, where the training and testing results of the model show an average error is around 5 %. Further, by applying a genetic algorithm (GA) to optimize the magnetic permeability of MRE, the theoretical optimal relative permeability curve of MRE is determined. To facilitate production, the goal is set to achieve 98 % of the maximum magnetic force, based on which further optimization is conducted to establish the engineering design range for MRE's relative permeability. Subsequently, MRE samples were designed and manufactured, and combined with a 36-pole EPM magnetic loading mechanism for experimental validation. The results show that the error between the predicted and experimental outcomes is 5.2 %, and the deep drawing experiments further confirm that the introduction of MRE helps the electronically controlled permanent magnet edging process to accommodate the deep drawing needs of slabs with varying thicknesses. This discovery provides valuable guidance for the improvement of the EPMBH process.

MoPPIt: De Novo Generation of Motif-Specific Binders with Protein Language Models.

The ability to precisely target specific motifs on disease-related proteins, whether conserved epitopes on viral proteins, intrinsically disordered regions within transcription factors, or breakpoint junctions in fusion oncoproteins, is essential for modulating their function while minimizing off-target effects. Current methods struggle to achieve this specificity without reliable structural information. In this work, we introduce a motif-specific PPI targeting algorithm, moPPIt, for de novo generation of motif-specific peptide binders from the target protein sequence alone. At the core of moPPIt is BindEvaluator, a transformer-based model that interpolates protein language model embeddings of two proteins via a series of multi-headed self-attention blocks, with a key focus on local motif features. Trained on over 510,000 annotated PPIs, BindEvaluator accurately predicts target binding sites given protein-protein sequence pairs with a test AUC > 0.94, improving to AUC > 0.96 when fine-tuned on peptide-protein pairs. By combining BindEvaluator with our PepMLM peptide generator and genetic algorithm-based optimization, moPPIt generates peptides that bind specifically to user-defined residues on target proteins. We demonstrate moPPIt's efficacy in computationally designing binders to specific motifs, first on targets with known binding peptides and then extending to structured and disordered targets with no known binders. In total, moPPIt serves as a powerful tool for developing highly specific peptide therapeutics without relying on target structure or structure-dependent latent spaces.

An Efficient Approach for Localizing Sensor Nodes in 2D Wireless Sensor Networks Using Whale Optimization-Based Naked Mole Rat Algorithm

Localization has emerged as an important and critical component of research in Wireless Sensor Networks (WSNs). WSN is a network of numerous sensors distributed across broad areas of the world to conduct numerous activities, including sensing the data and transferring it to various devices. Most applications, like animal tracking, object monitoring, and innumerable resources put in the interior as well as outdoor locations, need to identify the position of the occurring incident. The primary objective of localization is to identify the locality of sensor nodes installed in a network so that the location of a particular event can be traced. Different optimization approaches are observed in the work for solving the localization challenge in WSN and assigning the apt positions to undiscovered sensor nodes. This research employs the approach of localizing sensor nodes in a 2D platform utilizing an exclusive static anchor node and virtual anchors to detect dynamic target nodes by projecting these six virtual anchors hexagonally at different orientations and then optimizing the estimated target node co-ordinates employing Whale Optimization-based Naked Mole Rat Algorithm (WONMRA). Moreover, the effectiveness of a variety of optimization strategies employed for localization is compared to the WONMRA strategy concerning localization error and the number of nodes being localized, and it has been investigated that the average error in localization is 0.1999 according to WONMRA and is less than all other optimization techniques.

A particle swarm optimization-based deep clustering algorithm for power load curve analysis

To address the inflexibility of the convolutional autoencoder (CAE) in adjusting the network structure and the difficulty of accurately delineating complex class boundaries in power load data, a particle swarm optimization deep clustering method (DC-PSO) is proposed. First, a particle swarm optimization algorithm for automatically searching the optimal network architecture and hyperparameters of CAE (AHPSO) is proposed to obtain better reconstruction performance. Then, an end-to-end deep clustering model based on a reliable sample selection strategy is designed for the deep clustering algorithm to accurately delineate the category boundaries and further improve the clustering effect. The experimental results show that the DC-PSO algorithm exhibits high clustering accuracy and higher performance for the power load profile clustering.

HSHEP: An Optimization-Based Code Smell Refactoring Sequencing Technique

The process of refactoring enhances software quality by modifying its design composition while preserving its core framework. However, addressing code smells without appropriate prioritization can be ineffective. Code smells significantly increase maintenance costs and obstruct system evolution. Refactoring sequencing techniques mitigate these issues by improving a system’s internal structure without altering its external behavior. In large-scale systems, the sheer number of code smells can be overwhelming, and not all can be automatically resolved. Hence, prioritizing code smells based on criteria such as risk and importance is essential. This paper introduces a novel hybrid approach utilizing the hybrid spotted hyena and emperor penguin (HSHEP) optimization-based algorithm. This approach aims to optimize the sequence of code smell bugs by incorporating maintainer opinions and requirements, thereby maximizing the resolution of critical code smells. Unlike existing technologies, the HSHEP algorithm combines the strengths of two optimization strategies, offering a unique and innovative solution to refactoring challenges. To validate the effectiveness of the proposed method, it was applied to various large-scale open-source systems, analyzing five different types of code smells. Results demonstrated a significant improvement in maintenance efficiency and system evolution, confirming the superior performance and practical applicability of the HSHEP-based approach.

Output Reachable Set-Based Leader-Following Consensus of Positive Agents Over Switching Networks.

This work addresses the output reachable set-based leader-following consensus problem, focusing on a group of positive agents over directed dwell-time switching networks. Two types of non-negative disturbances, namely, 1) L1 -norm bounded disturbances and 2) L∞,1 -norm bounded disturbances are studied. Meanwhile, a class of directed dwell-time switching networks for modeling the communication protocol of positive agents is investigated. To deal with the presence of disturbances, an output-feedback control protocol is developed to achieve a robust consensus with positivity preserved based on the output reachable set. By exploiting the positive characteristics, switched linear copositive Lyapunov functions are adopted to establish output reachable set-based consensus conditions. These conditions can facilitate the control protocol design by solving a bilinear programming problem, and also generate hyperpyramidal regions to confine the output consensus error. A particle swarm optimization-based (PSO-based) algorithm is applied to compute the controller gains and optimize the volume of the hyperpyramids. The proposed methods are verified by the presented numerical case studies.

Techno-economic optimization and working fluid selection of a biogas-based dual-loop bi-evaporator ejector cooling cycle involving power-to-hydrogen and water facilities

Biogas-fueled decentralized energy systems featuring gas turbines emerge as pivotal players in the quest for more adaptable and eco-friendly energy solutions. This study presents the design of a cutting-edge multigeneration system that harnesses the potential of a gas turbine coupled with ejector-driven dual-loop bi-evaporator technology, reverse osmosis desalination, proton exchange membrane electrolysis, and an organic Rankine cycle. The research encompasses an extensive sensitivity analysis and deploys a genetic algorithm-based optimization technique. Through meticulous Optimization, the system achieves remarkable outcomes, including electricity generation, cooling capacity, heating capacity, desalinated water production, and hydrogen yield, measured at 957.3 kW, 231.4 kW, 272.3 kW, 7.336 kg/s, and 0.99 kg/h, respectively. Notably, this performance surpasses the base case by 2% and 8.9%, yielding exergy efficiency and total unit cost improvements of 33.3% and 16.4 $/GJ. Among the critical decisions made was selecting an organic working fluid for the organic Rankine cycle. Through rigorous evaluation, isobutene emerged as the optimal choice, demonstrating significantly improved output power compared to alternatives. Isobutane as the working fluid led to a substantial increase in overall exergy efficiency and a resultant total unit cost of 18.06 $/GJ, accompanied by an impressive 23.48% exergy efficiency.

Accurate Reconstruction of Multiple Basis Images Directly From Dual Energy CT Data.

We develop optimization-based algorithms to accurately reconstruct multiple ( 2) basis images directly from dual-energy (DE) data in CT. In medical and industrial CT imaging, some basis materials such as bone, metals, and contrast agents of interest are confined often spatially within regions in the image. Exploiting this observation, we develop an optimization-based algorithm to reconstruct, directly from DE data, basis-region images from which multiple ( 2) basis images and virtual monochromatic images (VMIs) can be obtained over the entire image array. We conduct experimental studies using simulated and real DE data in CT, and evaluate basis images and VMIs obtained in terms of visual inspection and quantitative metrics. The study results reveal that the algorithm developed can accurately and robustly reconstruct multiple ( 2) basis images directly from DE data. The developed algorithm can yield accurate multiple ( 2) basis images, VMIs, and physical quantities of interest from DE data in CT. The work may provide insights into the development of practical procedures for reconstructing multiple basis images, VMIs, and physical quantities from DE data in applications. The work can be extended to reconstruct multiple basis images in multi-spectral or photon-counting CT.

Manifold Optimization-Based Data Detection Algorithm for Multiple-Input–Multiple-Output Orthogonal Frequency-Division Multiplexing Systems under Time-Varying Channels

Recently, multiple-input–multiple-output (MIMO) orthogonal frequency-division multiplexing (OFDM) systems have gained significant attention in the field of wireless communications. The utilization of the Riemannian manifold has become prevalent in MIMO-OFDM systems. However, the existing data detection algorithms for MIMO-OFDM systems are mostly designed for block fading channels. Additionally, these algorithms often suffer from high computational complexity. In this paper, we propose a data detection algorithm on the basis of Riemannian manifold optimization for MIMO-OFDM systems under time-varying channels. The core concept of this algorithm is to optimize the transmitted signals by solving the manifold optimization problem in the case of time-varying channels. In order to reduce the computational complexity of the algorithm, we improve the proposed algorithm by dividing the transmitted signals into multiple subframes for solving the optimization problem separately and using the pilots to maintain the performance of the algorithm. In the simulation, the performance of multiple proposed algorithms and the forced-zero detection algorithm under different parameter settings are compared. The simulation results show that the proposed algorithm demonstrates good bit error rate and computational complexity performances.

Genetic Algorithm-Based Optimization of Clustering Algorithms for the Healthy Aging Dataset

Clustering is a crucial and, at the same time, challenging task in several application domains. It is important to incorporate the optimum feature finding into our clustering algorithms for better exploration of features and to draw meaningful conclusions, but this is difficult when there is no or little information about the importance or relevance of features. To tackle this task in an efficient manner, we employ the natural evolution process inherent in genetic algorithms (GAs) to find the optimum features for clustering the healthy aging dataset. To empirically verify the findings, genetic algorithms were combined with a number of clustering algorithms, including partitional, density-based, and agglomerative clustering algorithms. A variant of the popular KMeans algorithm, named KMeans++, gave the best performance on all performance metrics when combined with GAs.

Biological-equivalent-dose-based integrated optimization framework for fast-energy-switching Bragg peak FLASH-RT using single-beam-per-fraction.

When comparing the delivery of all beams per fraction (ABPF) to single beam per fraction (SBPF), it is observed that SBPF not only helps meet the FLASH dose threshold but also mitigates the uncertainty with beam switching in the FLASH effect. However, SBPF might lead to a higher biological equivalent dose in 2Gy (EQD2) for normal tissues. This study aims to develop an EQD2-based integrated optimization framework (EQD2-IOF), encompassing robust dose, delivery efficiency, and beam orientation optimization (BOO) for Bragg peak FLASH plans using the SBPF treatment schedule. The EQD2-IOF aims to enhance both dose sparing and the FLASH effect. A superconducting gantry was employed for fast energy switching within 27ms, while universal range shifters were utilized to improve beam current in the implementation of FLASH plans with five Bragg peak beams. To enhance dose delivery efficiency while maintaining plan quality, a simultaneous dose and spot map optimization (SDSMO) algorithm for single field optimization was incorporated into a Bayesian optimization-based auto-planning algorithm. Subsequently, a BOO algorithm based on Tabu search was developed to select beam angle combinations (BACs) for 10 lung cases. To simultaneously consider dose sparing and FLASH effect, a quantitative model based on dose-dependent dose modification factor (DMF) was used to calculate FLASH-enhanced dose distribution. The EQD2-IOF plan was compared to the plan optimized without SDSMO using BAC selected by a medical physicist (Manual plan) in the SBPF treatment schedule. Meanwhile, the mean EQD2 in the normal tissue was evaluated for the EQD2-IOF plan in both SBPF and ABPF treatment schedules. No significant difference was found in D2% and D98% of the target between EQD2-IOF plans and Manual Plans. When using a minimum DMF of 0.67 and a dose threshold of 4Gy, EQD2-IOF plans showed a significant reduction in FLASH-enhanced EQD2mean of the ipsilateral lung and normal tissue by 10.5% and 11.5%, respectively, compared to Manual plans. For normal tissues that received a dose greater than 70% of the prescription dose, using a minimum DMF of 0.7 for FLASH sparing compensated for the increase in EQD2mean resulting from replacing ABPF with SBPF schedules. The EQD2-IOF can automatically optimize SBPF FLASH-RT plans to achieve optimal sparing of normal tissues. With an energy switching time of 27ms, the loss of fractionate repairing using SBPF schedules in high-dose regions can be compensated for by the FLASH effect. However, when an energy switching time of 500ms is utilized, the SBPF schedule needs careful consideration, as the FLASH effect diminishes with longer irradiation time.

Turbo-RANS: straightforward and efficient Bayesian optimization of turbulence model coefficients

PurposeIndustrial simulations of turbulent flows often rely on Reynolds-averaged Navier-Stokes (RANS) turbulence models, which contain numerous closure coefficients that need to be calibrated. This paper aims to address this issue by proposing a semi-automated calibration of these coefficients using a new framework (referred to as turbo-RANS) based on Bayesian optimization.Design/methodology/approachThe authors introduce the generalized error and default coefficient preference (GEDCP) objective function, which can be used with integral, sparse or dense reference data for the purpose of calibrating RANS turbulence closure model coefficients. Then, the authors describe a Bayesian optimization-based algorithm for conducting the calibration of these model coefficients. An in-depth hyperparameter tuning study is conducted to recommend efficient settings for the turbo-RANS optimization procedure.FindingsThe authors demonstrate that the performance of the k-ω shear stress transport (SST) and generalized k-ω (GEKO) turbulence models can be efficiently improved via turbo-RANS, for three example cases: predicting the lift coefficient of an airfoil; predicting the velocity and turbulent kinetic energy fields for a separated flow; and, predicting the wall pressure coefficient distribution for flow through a converging-diverging channel.Originality/valueTo the best of the authors’ knowledge, this work is the first to propose and provide an open-source black-box calibration procedure for turbulence model coefficients based on Bayesian optimization. The authors propose a data-flexible objective function for the calibration target. The open-source implementation of the turbo-RANS framework includes OpenFOAM, Ansys Fluent, STAR-CCM+ and solver-agnostic templates for user application.

Flexi-BOPI: Flexible granularity pipeline inference with Bayesian optimization for deep learning models on HMPSoC

To achieve high-throughput deep learning (DL) model inference on heterogeneous multiprocessor systems-on-chip (HMPSoC) platforms, the use of pipelining for the simultaneous utilization of multiple resources has emerged as a promising solution. Nevertheless, current research faces two primary challenges: determining the optimal pipeline partitioning granularity, which directly influences the inference performance, and addressing the high time overhead of the search algorithms. To address these challenges, we propose Flexi-BOPI, a pipeline inference method for DL models of HMPSoCs. Flexi-BOPI offers flexible pipeline partitioning granularity down to a minimum size of a single core, enhancing the performance by better adapting to the diverse computational demands of different layers in DL models. Flexi-BOPI employs a Bayesian optimization-based search algorithm to significantly reduce the search overhead. In addition, we propose a surrogate model based on the heteroscedastic Gaussian process (HGP) to address the challenge of sample noise during the evaluation process. This approach can further reduce search overhead. Our experimental results demonstrate that the proposed method achieves significant improvements in inference performance and search overhead compared to existing methods.