Genetic Algorithm Research Articles

MemoriaNova: Optimizing Memory-Aware Model Inference for Edge Computing

In recent years, deploying deep learning models on edge devices has become pervasive, driven by the increasing demand for intelligent edge computing solutions across various industries. From industrial automation to intelligent surveillance and healthcare, edge devices are being leveraged for real-time analytics and decision-making. Existing methods face two challenges when deploying machine learning models on edge devices. The first challenge is handling the execution order of operators with a simple strategy, which can lead to a potential waste of memory resources when dealing with directed acyclic graph structure models. The second challenge is that they usually process operators of a model one by one to optimize the inference latency, which may lead to the optimization problem getting trapped in local optima. We present MemoriaNova, comprising BTSearch and GenEFlow, to solve these two problems. BTSearch is a graph state backtracking algorithm with efficient pruning and hashing strategies designed to minimize memory overhead during inference and enlarge latency optimization search space. GenEFlow, based on genetic algorithms, integrates latency modeling and memory constraints to optimize distributed inference latency. This innovative approach considers a comprehensive search space for model partitioning, ensuring robust and adaptable solutions. We implement BTSearch and GenEFlow and test them on eleven deep-learning models with different structures and scales. The results show that BTSearch can reach 12% memory optimization compared with the widely used random execution strategy. At the same time, GenEFlow reduces inference latency by 33.9% in distributed systems with four-edge devices.

X-Shooting ULLYSES: Massive Stars at low metallicity. IX.Empirical constraints on mass-loss rates and clumping parameters for OB supergiants in the Large Magellanic Cloud

Current implementations of mass loss for hot, massive stars in stellar evolution models usually include a sharp increase in mass loss when blue supergiants become cooler than $T_ eff 20-22$ kK. Such a drastic mass-loss jump has traditionally been motivated by the potential presence of a so-called bistability ionisation effect, which may occur for line-driven winds in this temperature region due to recombination of important line-driving ions. We perform quantitative spectroscopy using UV (ULLYSES program) and optical (XShootU collaboration) data for 17 OB-supergiant stars in the LMC (covering the range $T_ eff 14-32$ kK), deriving absolute constraints on global stellar, wind, and clumping parameters. We examine whether there are any empirical signs of a mass-loss jump in the investigated region, and we study the clumped nature of the wind. We used a combination of the model atmosphere code fastwind and the genetic algorithm (GA) code Kiwi-GA to fit synthetic spectra of a multitude of diagnostic spectral lines in the optical and UV. We find an almost monotonic decrease of mass-loss rate with effective temperature, with no signs of any upward mass loss jump anywhere in the examined region. Standard theoretical comparison models, which include a strong bistability jump thus severely overpredict the empirical mass-loss rates on the cool side of the predicted jump. Another key result is that across our sample we find that on average about 40<!PCT!> of the total wind mass seems to reside in the more diluted medium in between dense clumps. Our derived mass-loss rates suggest that for applications such as stellar evolution one should not include a drastic bistability jump in mass loss for stars in the temperature and luminosity region investigated here. The derived high values of interclump density further suggest that the common assumption of an effectively void interclump medium (applied in the vast majority of spectroscopic studies of hot star winds) is not generally valid in this parameter regime.

Performance Evaluation of Inerter and Negative Stiffness-Based Dampers for Seismic-Induced Vibration Response Control of a Cable-Stayed Bridge: A Comparative Study

The excessive main girder displacement responses of the large-span cable-stayed bridges during seismic events may precipitate collisions between the main girder and the approach bridge. The viscous dampers (VDs) have been extensively employed in the seismic response control of long-span bridges, yet their effectiveness is limited. To augment the seismic-induced vibration control performance of the VD, inerter element, negative stiffness (NS) element, and spring element have been incorporated into the VD to achieve energy dissipation capacity enhancement. The equations of motion for the simplified cable-stayed bridge-damper systems are established under stochastic seismic excitation, and the design parameters of inerter and NS-based damper are optimized by using the genetic algorithm. The seismic control performance of the VD and five types of dampers are systematically compared, demonstrating that these dampers can substantially enhance the main girder displacement mitigation performance in cable-stayed bridges. Owing to the NS characteristics attributes of the inerter element and NS element and the tuning effect of the spring element, the tuned viscous mass damper (TVMD) with NS achieves a 24.56% higher efficiency in control performance compared to the VD.

Research the lubrication characteristics on the modified main thrust roller of the slewing bearings

The load capacity and lubrication performance of the slewing bearing are fairly sensitive to edge effects at both ends of the main thrust rollers. Therefore, quantitative investigations of the edge effect at both ends of slewing bearing is practically important. In this study, an elastic fluid lubrication model of the slewing bearing that considered the external load and internal structural parameters is established. A modification function of rollers is included in a finite long-line contact elastic fluid lubrication model to perform the dynamic analysis. The simulated oil film’s thickness and pressure distributions are compared under different working conditions. And the edge effect is eliminated by modifying the rollers’ construction. The results indicate that under the large eccentric load, the speed of the slewing bearing must be decreased to ensure elastic fluid lubrication performance. A genetic algorithm that used to discover the optimal modification length shows the best busbar of the main thrust roller is 12 μm, which could decrease the vibration amplitude of the unmodified roller by 8.13%. This result gives a significant way for slewing bearing’s structural design.

Enhancing IoT Security Using GA-HDLAD: A Hybrid Deep Learning Approach for Anomaly Detection

The adoption and use of the Internet of Things (IoT) have increased rapidly over recent years, and cyber threats in IoT devices have also become more common. Thus, the development of a system that can effectively identify malicious attacks and reduce security threats in IoT devices has become a topic of great importance. One of the most serious threats comes from botnets, which commonly attack IoT devices by interrupting the networks required for the devices to run. There are a number of methods that can be used to improve security by identifying unknown patterns in IoT networks, including deep learning and machine learning approaches. In this study, an algorithm named the genetic algorithm with hybrid deep learning-based anomaly detection (GA-HDLAD) is developed, with the aim of improving security by identifying botnets within the IoT environment. The GA-HDLAD technique addresses the problem of high dimensionality by using a genetic algorithm during feature selection. Hybrid deep learning is used to detect botnets; the approach is a combination of recurrent neural networks (RNNs), feature extraction techniques (FETs), and attention concepts. Botnet attacks commonly involve complex patterns that the hybrid deep learning (HDL) method can detect. Moreover, the use of FETs in the model ensures that features can be effectively extracted from spatial data, while temporal dependencies are captured by RNNs. Simulated annealing (SA) is utilized to select the hyperparameters necessary for the HDL approach. In this study, the GA-HDLAD system is experimentally assessed using a benchmark botnet dataset, and the findings reveal that the system provides superior results in comparison to existing detection methods.

Analytical comparison of the boil-off gas recondensation process’s energy consumption and carbon emissions

With 28 operational liquefied natural gas receiving terminals in China by 2023, totaling a receiving capacity of 1.2 × 108 t/a, managing carbon emissions and boil-off gas has attracted increased attention. This study presents the optimization of 8 boil-off gas (BOG) recondensation processes, considering energy consumption and carbon emissions as assessment criteria, along with two types of vaporizers. The method involves analyzing how process parameters affect system performance using Hydraulic System Simulation Software, followed by optimizing system parameters with a genetic algorithm. It explores the impact of compressor output pressure, after-cooler inlet material ratio, and pre-cooler inlet material ratio on system energy consumption and carbon emissions. Results show that increased recondensing, compression, and cooling measures slightly elevate energy consumption but significantly reduce carbon emissions. Moreover, a higher BOG content enhances the carbon reduction effect. This study represents the first optimization of BOG recondensation processes focusing on carbon emissions, offering fresh theoretical insights and significant potential for enhancing energy efficiency and reducing carbon emissions.

High-throughput prediction of oral acute toxicity in Rat and Mouse of over 100,000 polychlorinated persistent organic pollutants (PC-POPs) by interpretable data fusion-driven machine learning global models

This study utilized available oral acute toxicity data in Rat and Mouse for polychlorinated persistent organic pollutants (PC-POPs) to construct data fusion-driven machine learning (ML) global models. Based on atom-centered fragments (ACFs), the collected high-throughput data overcame the applicability limitations, enabling accurate toxicity prediction for a wide range of PC-POPs series compounds using only single models. The data variances in the Rat training and test sets were 1.52 and 1.34, respectively, while for the Mouse, the values were 1.48 and 1.36, respectively. Genetic algorithm (GA) was used to build multiple linear regression (MLR) models and pre-screen descriptors, addressing the “black-box” problem prevalent in ML and enhancing model interpretability. The best ML models for Rat and Mouse achieved approximately 90 % prediction reliability for over 100,000 true untested compounds. Ultimately, a warning list of highly toxic compounds for eight categories of polychlorinated atom-centered fragments (PCACFs) was generated based on the prediction results. The analysis of descriptors revealed that dioxin analogs generally exhibited higher toxicity, because the heteroatoms and ring systems increased structural complexity and formed larger conjugated systems, contributing to greater oral acute toxicity. The present study provides valuable insights for guiding the subsequent in vivo tests, environmental risk assessment and the improvement of global governance system of pollutants.

Retraction Note: A novel design and performance improvement of symmetric multilevel inverter with reduced switches using genetic algorithm

Retraction Note: A novel design and performance improvement of symmetric multilevel inverter with reduced switches using genetic algorithm

Automatic Scheduling Method for Customs Inspection Vehicle Relocation Based on Automotive Electronic Identification and Biometric Recognition

This study presents an innovative automatic scheduling method for the relocation of customs inspection vehicles, leveraging Vehicle Electronic Identification (EVI) and biometric recognition technologies. With the expansion of global trade, customs authorities face increasing pressure to enhance logistics efficiency. Traditional vehicle scheduling often relies on manual processes and simplistic algorithms, resulting in prolonged waiting times and inefficient resource allocation. This research addresses these challenges by integrating EVI and biometric systems into a comprehensive framework aimed at improving vehicle scheduling. The proposed method utilizes genetic algorithms and intelligent optimization techniques to dynamically allocate resources and prioritize vehicle movements based on real-time data. EVI technology facilitates rapid identification of vehicles entering customs facilities, while biometric recognition ensures that only authorized personnel can operate specific vehicles. This dual-layered approach enhances security and streamlines the inspection process, significantly reducing delays. A thorough analysis of the existing literature on customs vehicle scheduling identifies key limitations in current methodologies. The automatic scheduling algorithm is detailed, encompassing vehicle prioritization criteria, dynamic path planning, and real-time driver assignment. The genetic algorithm framework allows for adaptive responses to varying operational conditions. Extensive simulations using real-world data from customs operations validate the effectiveness of the proposed method. Results indicate a significant reduction in vehicle waiting times—up to 30%—and an increase in resource utilization rates by approximately 25%. These findings demonstrate the potential of integrating EVI and biometric technologies to transform customs logistics management. Additionally, a comparison against state-of-the-art scheduling algorithms, such as NSGA-II and MOEA/D, reveals superior efficiency and adaptability. This research not only addresses pressing challenges faced by customs authorities but also contributes to optimizing logistics operations more broadly. In conclusion, the automatic scheduling method presented represents a significant advancement in customs logistics, providing a robust solution for managing complex vehicle scheduling scenarios. Future research directions will focus on refining the algorithm to handle peak traffic periods and exploring predictive analytics for enhanced scheduling optimization. Advancements in the intersection of technology and logistics aim to support more efficient and secure customs operations globally.

A 4IR-Driven operational risk model for CO2 storage in deepwater abandoned hydrocarbon reservoirs

This study presents the development of an advanced operational risk model that leverages Fourth Industrial Revolution (4IR) technologies to optimize carbon dioxide (CO2) storage within deepwater abandoned hydrocarbon reservoirs. The model systematically combines Artificial Neural Networks (ANN) with the optimization capabilities of Genetic Algorithms (GA) and the probabilistic analysis strengths of a Bayesian Network (BN) to perform dynamic and comprehensive risk assessments. By applying the model to a dataset covering a 200-year timeframe, it effectively forecasts CO2 storage capacities while simultaneously evaluating associated risks across different operational scenarios. One of the key innovations of this model is the introduction of a novel loss function designed to precisely manage forecast deviations and enhance the efficiency of operational processes. This function is critical in ensuring that the model remains robust and accurate in real-time risk assessments, allowing for more reliable decision-making in CO2 storage operations. In addition, the study conducts an economic evaluation that underscores the crucial role of 45Q tax credits in bolstering the financial sustainability of carbon sequestration projects. The analysis highlights how these credits significantly reduce the economic barriers to adopting carbon utilization, storage, and sequestration (CUSS) technologies, making large-scale implementation more feasible. The model's performance is underscored by its ability to achieve a 49% CO2 retention rate over two centuries, with an impressively low average error margin of 0.249%. These results highlight the model's impressive efficiency and accuracy, while also demonstrating its capacity to markedly improve the predictability of CO2 storage outcomes. The findings suggest that this model could play a pivotal role in advancing global sustainability efforts by optimizing CO2 storage processes, thereby contributing to the reduction of atmospheric CO2 levels and supporting long-term climate goals.

Lowering the PAPR of the optical OTFS-based 6G radio with a hybrid PTS-PSO genetic approach

Orthogonal Time Frequency Space (OTFS) is regarded as one of the best contenders for sixth-generation (6G) radio systems due to its ability to obtain excellent performance in high-speed scenarios. Peak-to-average power ratio (PAPR) significantly impacts and degrades the efficiency of the power amplifier used in OTFS-based 6G systems. The proposed article presents a genetic hybrid algorithm, Partial transmission sequence-particle swarm optimization (PTS-PSO), obtaining an optimal PAPR performance with low computation complexity. The PSO generates the best phase factor compared to the conventional phase generation method of PTS, making PTS-PSO more effective. The impacts such as PAPR, bit error rate (BER), power spectral density (PSD), and complexity of the proposed PTS-PSO are compared with the conventional selective mapping (SLM) and PTS methods for 64, 256, and 512 OTFS sub-carriers. The projected PTS-PSO attained a PAPR and PSD gain of 1–6 dB and − 540 while preserving the BER performance. The experimental results demonstrate that the projected PTS-PSO outperforms the conventional SLM and PTS algorithms with trivial intricacy.

The optimal design of structural stiffness for the CEPC vertex detector supports based on a multi-objective genetic algorithm

The optimal design of structural stiffness for the CEPC vertex detector supports based on a multi-objective genetic algorithm

Retraction Note: Enhancing low latency and high performance in clustered-based wireless sensor network routing using genetic algorithm for 5G networks

Retraction Note: Enhancing low latency and high performance in clustered-based wireless sensor network routing using genetic algorithm for 5G networks

Switchable Camouflage via Reflective Display

In nature, dynamic camouflage is performed by cephalopods and reptiles. Humans attempt to perform dynamic camouflage by employing display devices to show the surrounding background. In this work, a switchable camouflage device based on an electrophoretic display (EPD) is proposed. Color‐filter EPDs display colors by reflecting light through the color filters and black‐and‐white EPDs. The number of subpixels is found to be an important factor on color performance. To improve the poor saturation of color‐filter EPDs, the number of color filter subpixels is reduced. Compared with filters with three and four subpixels, a dual‐subpixel filter proposed in this work significantly improves the average saturation of red, green, and blue colors, with increases of 49% and 112%, respectively. Subsequently, the spectral characteristics of the color filter and black‐and‐white EPD are optimized by using genetic algorithm to reduce the average color difference between the display and the switchable target color, which can be reduced as low as 0.18. To visually demonstrate the color reproduction capability of the dual‐subpixel EPD, sample applications including the switchable vegetation and digital camouflages are designed and have a high degree of agreement with the background. In this work, an innovative and effective approach is introduced to dynamic camouflage.

Comparative Sensitivity Analysis of GA, PSO, and PSO-GA Algorithms for Carbon Environment Development and Sustainability

<p style="text-align:justify;"><span style="font-family:"Times New Roman",serif;" lang="EN-US">To effectively address the CO<sub>2</sub> emissions in the tourism place, in this study focuses a Sanshan scenic spot for precise prediction. The main factors contributing CO<sub>2</sub> emission as, accommodation, transportation and catering services which intends to improve the need of carbon emission forecasting methods. To address this issue, we proposed an innovatively integrating the optimization algorithms Genetic Algorithm and Particle Swarm Optimization for accurately predicting the CO<sub>2</sub> emissions. Initially, this system taken input from the CO2 emissions occurred from the year of 2010 to 2020. The optimization model contains the input variables, initializing populations, evaluating fitness values, and iterating until convergence to found best prediction model. The experimental results of the proposed PSO-GA system attains 9.86 highest sensitivity showcases best performance in predicting CO<sub>2 </sub>emissions, as evidenced by its superior Adjusted Rand Index value of 1 after 160 iterations. The predicted and measured values of CO<sub>2</sub> emissions using this algorithm remains significant and advanced carbon environment development at scenic spots. The PSO-GA was used to calculate the carbon emissions of three mountainous scenic spots over the next five years, and the medium and high carbon levels should change to a medium carbon development stage by 2024.</span></p>

Phase-transition metamaterial smart window for radiative cooling and privacy protection

This paper proposes a smart window for radiative cooling with adjustable transparency by leveraging the phase transition property of VO2. The proposed smart window exhibits daytime visible light transmission and near-infrared light reflection, all-day radiative cooling, and a nighttime privacy protection feature. It comprises a bottom layer of VO2/Ag/VO2 and top cubic two-dimensional SiO2 gratings, with structural parameters optimized by the genetic algorithm to ensure excellent optical performance. In the daytime, 75.4% visible light transmittance and 87.8% near-infrared reflectance can be achieved by utilizing the dielectric state of VO2. It also achieves low light transmission of 9.8% by using the metallic state of VO2, which is suitable for nighttime privacy protection. Furthermore, it has an all-day outside emissivity of 98.2% for radiative cooling, together with a low inside emissivity of 1.9% for effectively inhibiting the radiation heat transfer. In addition, the proposed structure is insensitive to the angles of incidence and the polarization of light, making it advantageous for radiative cooling. During the daytime, it reduces the temperature by 17.7 K compared to a glass of equal thickness when the non-radiative heat coefficient is 12 W/m2/K. At night, it achieves a cooling power of 124.7 W/m2, achieving a cooling effect of 8.9 K below ambient temperature. The proposed smart window is promising for various application scenarios for radiative cooling and privacy protection and could be used as windows for buildings and vehicles.

Optimizing production planning and sequencing in hot strip mills: an approach using multi-objective genetic algorithms

Optimizing production planning and sequencing in hot strip mills: an approach using multi-objective genetic algorithms

Leveraging Machine Learning to Optimize Pricing and Restocking Strategies for Fresh Vegetables

In the competitive and diverse retail market, restocking and pricing perishable vegetables is challenging. Using big data and machine learning to analyze sales data, this research can build predictive models to refine these strategies and predict trends[1][2]. This research analyzed sales data from 2020 to 2023, identifying key trends in sales volumes, seasonality, and consumer preferences. It incorporated these findings into a multi-objective programming model that also considered wholesale and loss rates. This research used statistical tests like the Kruskal-Wallis and Mann-Whitney U to detect significant sales differences across vegetable categories, while the Spearman correlation showed strong seasonal and culinary impacts on consumer buying behaviors. To optimize freshness and maximize sales and profits, this research utilized genetic algorithms and regression analyses, including an effective XGBoost model. This model had a Mean Absolute Percentage Error of 16.97 and an R² of 0.94, aiding in predicting daily sales and improving pricing strategies and restocking schedules. These methods helped enhance decision-making, ensuring vegetable freshness and maximizing profitability, with projected profits reaching ¥41,582.87. This study not only highlights the effectiveness of integrating big data and machine learning in the retail sector but also sets a benchmark for operational efficiency in managing perishable goods.

Traffic Congestion Scheduling for Underground Mine Ramps Based on an Improved Genetic Scheduling Algorithm

The dispatching of trackless transportation on the ramp of underground metal mines is closely related to the transportation efficiency of daily production equipment, personnel, and construction materials in the mine. The current dispatching of trackless transportation on the ramp of underground metal mines is discontinuous and imprecise, with unscientific vehicle arrangement leading to low efficiency and transportation congestion. This paper presents this study, which puts forward a kind of trackless transportation optimization method that can fully make use of the ramp in the roadway, and the slow slope fork point can be used for the trackless transportation vehicle passing section to improve the efficiency of trackless transportation on the ramp. This study adopts the principles of fuzzy logic and uses interval-based positioning instead of real-time positioning to effectively reduce the spatial complexity inherent in the algorithm. At the same time, this research presents a modified genetic algorithm that incorporates a time-loss fitness calculation. This innovation makes it possible to differentiate traffic priorities between different types of vehicles, thus bringing the scheduling scheme more in line with the economic objectives of the mining operations. Various parameters were determined and several sets of simulation experiments were carried out on the response speed and scheduling effect of the scheduling system, resulting in a 10 to 20 percent improvement for different vehicles in the efficiency of underground mining transport operations.

Crash resistance analysis and optimization of sponge like thin-walled pipes under multiple working conditions

The crashworthiness of bionic thin-walled structures has been the subject of considerable research interest. In this article, we emulate the bi-diagonal skeletal structure of deep-sea glass sponges and construct sponge-like multicellular thin-walled tubes (SLMTT) with a parameterized ratio of the number of two distinct single-celled organisms (CSS I and CSS II). Thin-walled sponge-like tubes of varying dimensions were manufactured using selective laser melting (SLM) technology. The crashworthiness of these structures was evaluated under axial compression, transverse compression, and three-point bending conditions through experimental and computational approaches, including parametric analyses of the inner and outer wall thicknesses. The results show that SLMTT I has better crashworthiness under the three working conditions, and the SEA of BLMESI is 40.9 and 26.13% higher than that of BLMESII. under transverse compression and three-point bending conditions, respectively. Finally, a multi-objective optimization design of SLMTT I under multiple operating conditions is carried out by the second generation of Non-dominated Sorting Genetic Algorithm (NSGA-II) with the aim of obtaining the ideal structure with maximum specific energy absorption and minimum initial peak crushing force. This study offers novel insights into the design of bionic energy-absorbing structures under diverse operational scenarios.