Search

Neural networks (NNs) have revolutionized various fields, including aeronautics where it is applied in computational fluid dynamics, finite element analysis, load prediction, and structural optimization. Particularly in optimization, neural networks and deep neural networks are extensively employed to enhance the efficiency of genetic algorithms because, with this tool, it is possible to speed up the finite element analysis process, which will also speed up the optimization process. The main objective of this paper is to present how neural networks can help speed up the process of optimizing the geometries and composition of composite structures (dimension, topology, volume fractions, reinforcement architecture, matrix/reinforcement composition, etc.) compared to the traditional optimization methods. This article stands out by showcasing not only studies related to aeronautics but also those in the field of mechanics, emphasizing that the underlying principles are shared and applicable to both domains. The use of NNs as a surrogate model has been demonstrated to be a great tool for the optimization process; some studies have shown that the NNs are accurate in their predictions, with an MSE of 1×10−5 and MAE of 0.007%. It has also been observed that its use helps to reduce optimization time, such as up to a speed 47.5 times faster than a full aeroelastic model.

In today’s tech world of digitalization, engineers are leveraging tools such as artificial intelligence for analyzing data in order to enhance their capability in evaluating product quality effectively. This research study adds value by applying algorithms and various machine learning techniques—such as support vector regression, Gaussian process regression, and artificial neural networks—on a dataset related to the grinding process of UNS S34700 steel. What sets this study apart is its consideration of factors like three types of grinding wheels, four distinct cooling solutions, and seven varied depths of cut. These parameters are assessed for their impact on surface roughness and grinding forces, resulting in the conversion of information into insights. A relational equation with 25 coefficients is developed, using optimized algorithms to predict surface roughness with an 85 percent accuracy and grinding forces with a 90 percent accuracy rate. Learning from machine models like the Gaussian process regression exhibited stability, with an R2 value of 0.98 and a mean accuracy of 93 percent. Artificial neural networks achieved an R2 value of 0.96, and an accuracy rate of 90 percent. These findings suggest that machine learning techniques are versatile and precise when dealing with datasets. They align well with digitalization and predictive trends. In conclusion; machine learning provides flexibility and superior accuracy for predicting data trends compared to the formulaic approach, which is contained to existing datasets only. The versatility of machine learning highlights its significance in engineering practices for making data-informed decisions.

The global community is awaiting the advent of a self-driving vehicle that is safe, reliable, and capable of navigating a diverse range of road conditions and terrains. This requires a lot of research, study, and optimization. Thus, this work focused on implementing, training, and optimizing a convolutional neural network (CNN) model, aiming to predict the steering angle during driving (one of the main issues). The considered dataset comprises images collected inside a car-driving simulator and further processed for augmentation and removal of unimportant details. In addition, an innovative data-balancing process was previously performed. A CNN model was trained with the dataset, conducting a comparison between several different standard optimizers. Moreover, evolutionary optimization was applied to optimize the model’s weights as well as the optimizers themselves. Several experiments were performed considering different approaches of genetic algorithms (GAs) along with other optimizers from the state of the art. The obtained results demonstrate that the GA is an effective optimization tool for this problem.

In an efficient aerial package delivery scenario carried out by multiple Unmanned Aerial Vehicles (UAVs), a task allocation problem has to be formulated and solved in order to select the most suitable assignment for each delivery task. This paper presents the development methodology of an evolutionary-based optimization framework designed to tackle a specific formulation of a Drone Delivery Problem (DDP) with charging hubs. The proposed evolutionary-based optimization framework is based on a double-chromosome task encoding logic. The goal of the algorithm is to find optimal (and feasible) UAV task assignments such that (i) the tasks’ due dates are met, (ii) an energy consumption model is minimized, (iii) re-charge tasks are allocated to ensure service persistency, (iv) risk-aware flyable paths are included in the paradigm. Hard and soft constraints are defined such that the optimizer can also tackle very demanding instances of the DDP, such as tens of package delivery tasks with random temporal deadlines. Simulation results show how the algorithm’s development methodology influences the capability of the UAVs to be assigned to different tasks with different temporal constraints. Monte Carlo simulations corroborate the results for two different realistic scenarios in the city of Turin, Italy.

This research focuses on integrating seakeeping indices into the trajectory planning of a mother ship in order to minimize risks during UAV (unmanned aerial vehicle) takeoff and landing in challenging sea conditions. By considering vessel dynamics and environmental factors, the proposed trajectory planning algorithm computes optimal paths that prioritize the stability and safety of the ship, mitigating the impact of adverse weather on UAV operations. Specifically, the new adaptive weather routing model presented is based on a genetic algorithm. The model uses the previously evaluated response amplitude operators (RAOs) for the reference ship at different velocities and encounter angles, along with weather forecast data provided by the global wave model (GWAM). Preliminary evaluations confirm the effectiveness of the presented model in significantly improving the reliability of autonomous UAV operations from a mother ship across all encountered sea state conditions, particularly when compared with a graph-based solution. The current results clearly demonstrate that it is possible to achieve appreciable improvements in ship seakeeping performance, thereby making UAV-related operations safer.

The increasing number, frequency, and sophistication of phishing website-based attacks necessitate the development of robust solutions for detecting phishing websites to enhance the overall security of cyberspace. Drawing inspiration from natural processes, nature-inspired metaheuristic techniques have been proven to be efficient in solving complex optimization problems in diverse domains. Following these successes, this research paper aims to investigate the effectiveness of metaheuristic techniques, particularly Genetic Algorithms (GAs), Differential Evolution (DE), and Particle Swarm Optimization (PSO), in optimizing the hyperparameters of machine learning (ML) algorithms for detecting phishing websites. Using multiple datasets, six ensemble classifiers were trained on each dataset and their hyperparameters were optimized using each metaheuristic technique. As a baseline for assessing performance improvement, the classifiers were also trained with the default hyperparameters. To validate the genuine impact of the techniques over the use of default hyperparameters, we conducted statistical tests on the accuracy scores of all the optimized classifiers. The results show that the GA is the most effective technique, by improving the accuracy scores of all the classifiers, followed by DE, which improved four of the six classifiers. PSO was the least effective, improving only one classifier. It was also found that GA-optimized Gradient Boosting, LGBM and XGBoost were the best classifiers across all the metrics in predicting phishing websites, achieving peak accuracy scores of 98.98%, 99.24%, and 99.47%, respectively.

The present research focuses on optimizing 5G base station deployment and visualization, addressing the escalating demands for high data rates and low latency. The study compares the effectiveness of Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Simulated Annealing (SA), and Grey Wolf Optimizer (GWO) in both Urban Macro (UMa) and Remote Macro (RMa) deployment scenarios that overcome the limitations of the current method of 5G deployment, which involves adopting Non-Standalone (NSA) architecture. Emphasizing population density, the optimization process eliminates redundant base stations for enhanced efficiency. Results indicate that PSO and GA strike the optimal balance between coverage and capacity, offering valuable insights for efficient network planning. The study includes a comparison of 28 GHz and 3.6 GHz carrier frequencies for UMa, highlighting their respective efficiencies. Additionally, the research proposes a 2.6 GHz carrier frequency for Remote Macro Antenna (RMa) deployment, enhancing 5G Multi-Tier Radio Access Network (RAN) planning and providing practical solutions for achieving infrastructure reduction and improved network performance in a specific geographical context.

Monolithic reinforced concrete floor slabs are one of the most common types of building structures, and their optimization is an urgent task. The article presents the methodology for finding the optimal position of point supports under a reinforced concrete floor slab of arbitrary configuration at arbitrary load. The slab is considered thin, elastic and isotropic, with constant over-the-area stiffness, that is, the reinforcement is not taken into account or is constant. The solution is performed using the finite element method in combination with the nonlinear optimization methods. Finite element analysis is implemented by authors in MATLAB (R2024a) environment in such a way that the location of the columns may not coincide with the nodes of the finite element mesh of the slab. This allows to significantly increase the efficiency of solving the optimization problem compared to previously used algorithms, including the Monte Carlo method. Boundary conditions are taken into account using the Lagrange multiplier method. As an optimization criterion, the maximum deflection value is used, as well as the value of the potential strain energy. The effectiveness of six nonlinear optimization methods is compared in the example of a square slab under the action of a uniformly distributed load. For solutions obtained using the pattern search, simulated annealing and internal point methods, the maximum deflections are at least 1.2 times higher than for solutions obtained using the particle swarm method and genetic algorithm. An example of real object optimization is also presented. By changing the position of seven columns, it was possible to reduce the maximum deflection of the floor slab by 1.6 times.

This study presents a methodology to optimize concrete mixtures by integrating machine learning (ML) and genetic algorithms. ML models are used to predict compressive strength, while genetic algorithms optimize the mixture cost under quality constraints. Using a dataset of over 19,000 samples from a local ready-mix concrete producer, various predictive ML models were trained and evaluated regarding cost-effective solutions. The results show that the optimized mixtures meet the desired compressive strength range and are cost-efficient, thus having 50% of the solutions yielding a cost below 98% of the test cases. CatBoost emerged as the best ML technique, thereby achieving a mean absolute error (MAE) below 5 MPa. This combined approach enhances quality, reduces costs, and improves production efficiency in concrete manufacturing.

Artificial intelligence paradigms hold significant potential to advance nanophotonics. This study presents a novel approach to designing a plasmonic absorber using an artificial neural network as a surrogate model in conjunction with a genetic algorithm. The methodology involved numerical simulations of multilayered metal–dielectric plasmonic structures to establish a dataset for training an artificial neural network (ANN). The results demonstrate the proficiency of the trained ANN in predicting reflectance spectra and its ability to generalize intricate relationships between desired performance and geometric configurations, with values of correlation higher than 98% in comparison with ground-truth electromagnetic simulations. Furthermore, the ANN was employed as a surrogate model in a genetic algorithm (GA) loop to achieve target optical behaviors. The proposed methodology provides a powerful means of inverse designing multilayered metal–dielectric devices tailored for visible band wavelength filtering. This research demonstrates that the integration of AI-driven approaches in nanophotonics leads to efficient and effective design strategies.

Numerous conflicting objectives exist in the engineering field, and resolving these conflicts to reduce costs constitutes a problem that demands top-priority consideration. A model for tool wear and a multi-quadratic regression model for milling forces were developed to accurately predict the trends of wear on the rake face of the milling tool and the variations in milling forces. The influence of milling parameters (spindle speed, n; feed rate, vf; axial milling depth, ap) on both the wear of the rake face and milling force was analyzed by means of orthogonal experiments. The findings indicated that the impact of these parameters on the wear ranked in the following order: n > vf > ap. In contrast, for milling force, F, the ranking was ap > vf > n. Utilizing MATLAB’s genetic algorithm, an optimization procedure was conducted with multiple objectives including the wear of the rake face, milling force, and material removal rate; subsequently, a Pareto optimal solution set was generated for milling parameters based on practical processing requirements.

To establish a finite element model that accurately represents the dynamic characteristics of actual super high-rise building and improve the accuracy of the finite element simulation results, a finite element model updating method for super high-rise building is proposed based on the response surface method (RSM). Taking a 120 m super high-rise building as the research object, a refined initial finite element model is firstly established, and the elastic modulus and density of the main concrete and steel components in the model are set as the parameters to be updated. A significance analysis was conducted on 16 parameters to be updated including E1–E8, D1–D8, and the first 10 natural frequencies of the structure, and 6 updating parameters are ultimately selected. A sample set of updating parameters was generated using central composite design (CCD) and then applied to the finite element model for calculation. The response surface equations for the first ten natural frequencies were obtained through quadratic polynomial fitting, and the optimal solution of the objective function was determined using a genetic algorithm. The results of the engineering case study indicate that the errors in the first ten natural frequencies of the updated finite element model are all within 5%. The updated model accurately reflects the current situation of the super high-rise building and provides a basis for super high-rise building health monitoring, damage detection, and reliability assessment.

To address the challenges posed by the large scale of agricultural machinery trajectory data and the complexity of actual movement trajectories, this paper proposes a two-stage joint clustering method for agricultural machinery trajectory recognition to enhance accuracy and robustness. The first stage involves trajectory clustering, where the spatial distribution characteristics of agricultural machinery trajectories are analyzed, and the position coordinates and the number of neighboring points of trajectory points are extracted as features. The silhouette coefficient method is used to determine the optimal number of clusters k for the K-Means algorithm, thus reducing the data scale. The second stage focuses on trajectory recognition, where a list of Eps and Minpts parameters is generated based on the statistical properties of the trajectory dataset. The Genetic Algorithm is employed for parameter optimization to determine the optimal DBSCAN parameters, enabling precise identification of field operation trajectories and road travel trajectories. Experimental results show that this method achieves mean values of 91.55% for Accuracy, 95.41% for Precision, 89.86% for Recall, and 92.41% for F1-score on a sample dataset of 337 trajectories, representing improvements of 12.8%, 5.13%, 7.79%, and 6.84%, respectively, over the traditional DBSCAN algorithm. Additionally, the Runtime of the two-stage joint clustering method is approximately 30% shorter than that of single-stage clustering. Compared with mainstream deep learning models such as LSTM and Transformer, this method delivers comparable recognition accuracy without the need for labeled data training, significantly reducing recognition costs. The proposed method achieves accurate and robust recognition of agricultural machinery trajectories and holds broad application potential in practical scenarios.

Water delivery management in China’s irrigation districts has traditionally prioritized the main canal system, often overlooking the water-saving potential of the final canals and field irrigation, which offer substantial opportunities to enhance water use efficiency and conserve agricultural water resources. This study summarizes and defines the integrated water management of final canals and field irrigation as terminal water management. An optimization method was developed to improve terminal water management, which includes optimizing irrigation quotas based on water balance and scheduling final canal water delivery to minimize seepage losses. A genetic algorithm was employed to solve the problem. The method was applied to the Hongjin irrigation district in Jiangsu Province, China. In 2020, paddy water management was observed, revealing that the irrigation amount for organic and traditional rice was 1113 mm and 956 mm, respectively. Conventional irrigation and water delivery practices have led to extensive drainage, significant rainwater wastage, and inefficient water use. The optimized irrigation quotas for organic and traditional rice resulted in water savings of 302.5 mm and 325.9 mm, respectively, compared to the 2020 monitored data. An irrigation event in early August during a 75% hydrological frequency year was selected as an example. With conventional scheduling, optimized final canal water delivery scheduling reduced the seepage losses from 6.3% to 4.6%, shortened the irrigation time from 17 h to 14 h, and stabilized canal flow rates. The proposed optimization method is a valuable tool for enhancing terminal water management and supporting better irrigation decisions in irrigation districts.

This study explored the optimization of groupage intercity delivery using a two-stage genetic algorithm (GA) framework, developed with the BaumEvA Python library. The primary objective was to minimize the transportation costs by strategically positioning regional branch warehouses within a logistics network. In the first stage, the GA selected optimal branch warehouse locations from a set of candidate cities. The second stage addressed the vehicle routing problem (VRP) by employing a combinatorial GA to optimize the delivery routes. The GA framework was designed to minimize the total costs associated with intercity and last-mile deliveries, factoring in warehouse locations, truck routes, and vehicle types for last-mile fulfillment while ensuring capacity constraints are adhered to. By solving both line haul and last-mile delivery subproblems, this solution adjusted variables related to warehouse placement, cargo volumes, truck routing, and vehicle selection. The integration of such optimization techniques into the logistics workflow allowed for streamlined operations and reduced costs.

Due to climate change, enhancing outdoor and indoor thermal comfort is increasingly important. Solar radiation drives temperature increases, making solar exposure reduction essential in urban design. Most previous research has focused on parametric analysis to optimize small urban blocks, often overlooking the impact of the overall urban district (UD) on reducing Solar Radiation Access (SRA). This work aims to find the optimized UD to minimize SRA and maximize Floor Area (FA). The proposed methodology is developed to achieve these objective functions using a single-objective Genetic Algorithm (GA) with three street layout patterns: random, radial, and grid layout. Further SRA analysis is conducted at the urban block level, focusing on blocks with the highest SRA in the optimized UD to achieve further SRA reduction while maintaining the same FA. Dubai Silicon Oasis district in the UAE was selected as a case study. Elk2-0.3.1 (GIS data), Ladybug (1.7.0), DeCodingSpaces-Toolbox (2021.10), and Galapagos (1.0.0007) Plugins in Grasshopper (0.9.0076) were used. The results show that the radial street pattern achieved better results with an 8.4% reduction in SRA with an 8.9% increase in FA. Additional analysis of the blocks with the highest SRA can achieve an additional 7.4% reduction in SRA.

The shape of the blade strongly influences the aerodynamic behavior of wind turbines; therefore, it is essential to optimize its design to maximize the energy harvested from the wind. Some works address this optimized design problem using CFD, a tool that requires a lot of computational resources and time and starts from scratch. This work describes a new automated design method to generate aerodynamic profiles of wind turbines using existing blades as a base, which speeds up the design process. The optimization is performed using heuristic techniques, and the aim is to improve the characteristics of the blade shape which impact resilience and durability. Specifically, the glide ratio is maximized to capture maximum energy while ensuring specific design parameters, such as maximum thickness or optimal angle of attack. This methodology can obtain results more quickly and with lower computational cost, in addition to integrating these two design parameters into the optimization process, aspects that have been largely neglected in previous works. The analytical model of the blades is described by a class of two-dimensional shapes suitable for representing airfoils. The drag and lift coefficients are estimated, and a metaheuristic optimization technique, genetic algorithm, is applied to maximize the glide ratio while reducing the difference from the desired design parameters. Using this methodology, three new airfoils have been generated and compared with the existing starting models, S823, NACA 2424, and NACA 64418, achieving improvements in the maximum lift and maximum glide ratio of up to 13.8% and 39%, respectively. For validation purposes, a small 10 kW horizontal-axis wind turbine is simulated using the best design of the blades. The comparison with the existing blades focuses on the calculation of the generated power, the power coefficient, torque, and torque coefficient. For the new airfoils, improvements of 6.7% in the power coefficient and 5.5% in the torque coefficient were achieved. This validates the methodology for optimizing the blade airfoils.

The selection of construction sites for Cultural and Museum Public Buildings (CMPBs) has a profound impact on their future operations and development. To enhance site selection and planning efficiency, we developed a predictive model integrating Artificial Neural Networks (ANNs) and Genetic Algorithms (GAs). Taking Shanghai as our case study, we utilized over 1.5 million points of interest data from Amap Visiting Vitality Values (VVVs) from Dianping and Shanghai’s administrative area map. We analyzed and compiled data for 344 sites, each containing 39 infrastructure data sets and one visit vitality data set for the ANN model input. The model was then tested with untrained data to predict VVVs based on the 39 input data sets. We conducted a multi-precision analysis to simulate various scenarios, assessing the model’s applicability at different scales. Combining GA with our approach, we predicted vitality improvements. This method and model can significantly contribute to the early planning, design, development, and operational management of CMPBs in the future.

Aiming at some shortcomings of the genetic algorithm to solve the path planning in a global static environment, such as a low efficiency of population initialization, slow convergence speed, and easy-to-fall-into the local optimum, an improved genetic algorithm is proposed to solve the path planning problem. Firstly, the environment model is established by using the grid method; secondly, in order to overcome the difficulty of a low efficiency of population initialization, a population initialization method with directional guidance is proposed; finally, in order to balance the global and local optimization searching and to speed up the solution speed, the proposed non-common point crossover operator, range mutation operator, and simplification operator are used in combination with the one-point crossover operator and one-point mutation operator in the traditional genetic algorithm to obtain an improved genetic algorithm. In the simulation experiment, Experiment 1 verifies the effectiveness of the population initialization method proposed in this paper. The success rates in Map 1, Map 2, Map 3, and Map 4 were 56.3854%, 55.851%, 34.1%, and 24.1514%, respectively, which were higher than the two initialization methods compared. Experiment 2 verifies the effectiveness of the genetic algorithm (IGA) improved in this paper for path planning. In four maps, the path planning is compared with the five algorithms and the shortest distance is achieved in all of them. The two experiments show that the improved genetic algorithm in this paper has advantages in path planning.

Regarding autonomous vehicle navigation, reinforcement learning is a technique that has demonstrated significant results. Nevertheless, it is a technique with a high number of parameters that need to be optimised without prior information, and correctly performing this is a complicated task. In this research study, a system based on the principles of reinforcement learning, specifically on the concept of rewards, is presented. A mathematical expression was proposed to control the vehicle’s direction based on its position, the obstacles in the environment and the destination. In this equation proposal, there was only one unknown parameter that regulated the degree of the action to be taken, and this was optimised through the genetic algorithm. In this way, a less computationally expensive navigation algorithm was presented, as it avoided the use of neural networks. The controller’s time to obtain the navigation instructions was around 6.201·10−4 s. This algorithm is an efficient and accurate system which manages not to collide with obstacles and to reach the destination from any position. Moreover, in most cases, it has been found that the proposed navigations are also optimal.

Airborne array synthetic aperture radar (SAR) can achieve three-dimensional (3D) imaging of the observed scene in a single flight. Nevertheless, the imaging process of airborne array SAR is subject to various parameter errors due to unstable factors. Such errors degrade the quality of 3D imaging, particularly for the elevation imaging results, which necessitates the employment of super-resolution algorithms. The most significant error parameters include the amplitude and phase imbalances between multiple channels, as well as the phase-center positions of each channel. Owing to the coupled nature of these parameter errors, the calibration accuracy for each parameter independently is relatively sub-par, while super-resolution algorithms have strict demands for parameter precision. Addressing these challenges, this article proposes a multi-parameter calibration method for airborne array SAR based on the Newton method and the genetic algorithm. Initially, a least squares model for multi-parameter calibration is established, followed by leveraging the global optimization characteristics of genetic algorithms and the rapid convergence property of the Newton method. The genetic algorithm is utilized to locate a sub-optimal solution in proximity to the optimal one, subsequently converging swiftly to the optimal solution via the Newton method, which incorporates second-order information. This approach averts the pitfalls of local convergence due to large initial value errors, thereby enhancing the algorithm’s robustness. The proposed method effectively enhances the precision of multi-parameter calibration, which is of significant importance in ensuring the quality of 3D imaging of airborne array SAR.

This study introduces a two-stage optimization method for designing an ultrawideband polarization conversion metasurface. By integrating topology and shape optimization techniques, the proposed method expands the design space to achieve enhanced polarization conversion bandwidth. The first stage employs genetic-algorithm-based topology optimization to establish the initial structural configuration through binary coding. Subsequently, the second stage refines the design through shape optimization by extracting and modifying the boundaries of the topology-optimized structure. The optimized design demonstrates high polarization conversion efficiency (>90%) across 4.08–14.39 GHz, yielding a relative bandwidth of 111.64%, which represents a 4.88% improvement compared to topology-only optimization. This enhancement demonstrates the effectiveness of our combined optimization method in ultrawideband polarization conversion metasurface design, offering a promising method for developing high-performance electromagnetic devices.

Recently, laser-induced coloring of metal surfaces has emerged as a hot topic in the field of color manufacturing. In existing research, we have not been able to find a reliable method to swiftly acquire all the color ranges achievable with current materials. This limitation hinders further research and application of laser-induced metal coloring, making it challenging to scientifically and effectively reproduce colors in images. In this study, we introduced a genetic algorithm tailored for predicting the maximization of color gamut area. By employing an elitist strategy for genetic selection and predicting the maximum color gamut among a multi-objective optimization parameter population, we successfully explored the color gamut of stainless steel. The color gamut S converged to 0.0022, offering a rapid and efficient approach for color gamut exploration. Building on this, we devised a comprehensive image color reproduction process and developed an image color gamut mapping toolkit and an image vectorization toolkit. These tools are designed for color separation, color gamut mapping, and vectorization of target images, enabling successful color reproduction through laser-induced coloring. Additionally, we conducted a color difference analysis experiment using 2 mm 304 stainless steel, demonstrating that material thickness can mitigate color cast issues in laser-induced coloring. The color difference (ΔE) values in the color reproduction experiment were 2.18, 2.97, and 2.72, respectively, indicating the reliability of image color reproduction on stainless steel surfaces. This research addresses the challenge of color gamut exploration in laser-induced coloring, presenting a novel solution for color reproduction via laser-induced coloring on metal surfaces, and holds promising applications.

Pressure swing adsorption (PSA) technology is among the most efficient techniques for purifying and separating hydrogen. A layered adsorption bed composed of activated carbon and zeolite 5A for a gas mixture (H2: 56.4 mol%, CH4: 26.6 mol%, CO: 8.4 mol%, N2: 5.5 mol%, CO2: 3.1 mol%) PSA model was built. The simulation model was validated using breakthrough curves. Then, a six-step PSA cycle model was built, and the purification performance was studied. The Box–Behnken design (BBD) method was utilized in Design Expert software (version 10) to optimize the PSA purification performance. The independent optimization parameters included the adsorption time, the pressure equalization time, and the feed flow rate. Quadratic regression models can be derived to represent the responses of purity and productivity. To explore a better optimization solution, a novel optimization method using machine learning with a back propagation neural network (BPNN) was then proposed, and a kind of heuristic algorithm–genetic algorithm (GA) was introduced to enhance the architecture of the BPNN. The predicted outputs of hydrogen production using two kinds of models based on the BPNN–GA and the BBD method integrated with the BPNN–GA (BBD–BPNN–GA). The findings revealed that the BBD–BPNN–GA model exhibited a mean square error (MSE) of 0.0005, with its R–value correlation coefficient being much closer to 1, while the BPNN–GA model exhibited an MSE of 0.0035. This suggests that the BBD–BPNN–GA model has a better performance, as evidenced by the lower MSE and higher correlation coefficient compared to the BPNN–GA model.

Efficient path planning is vital for multi-UAV inspection missions, yet the comparative effectiveness of different optimization strategies has not received much attention. This paper introduces the first application of the Genetic Algorithm (GA) and Hill Climbing (HC) to multi-UAV inspection of indoor pipelines, providing a unique comparative analysis. GA exemplifies the global search strategy, while HC illustrates an enhanced stochastic local search. This comparison is impactful as it highlights the trade-offs between exploration and exploitation—two key challenges in multi-UAV path optimization. It also addresses practical concerns such as workload balancing and energy efficiency, which are crucial for the successful implementation of UAV missions. To tackle common challenges in multi-UAV operations, we have developed a novel repair mechanism. This mechanism utilizes problem-specific repair heuristics to ensure feasible and valid solutions by resolving redundant or missed inspection points. Additionally, we have introduced a penalty-based approach in HC to balance UAV workloads. Using the Crazyswarm simulation platform, we evaluated GA and HC across key performance metrics: energy consumption, travel distance, running time, and maximum tour length. The results demonstrate that GA achieves a 22% reduction in travel distance and a 23% reduction in energy consumption compared to HC, which often converges to suboptimal solutions. Additionally, GA outperforms HC, Greedy, and Random strategies, delivering at least a 13% improvement in workload balancing and other metrics. These findings establish a novel and impactful benchmark for comparing global and local optimization strategies in multi-UAV tasks, offering researchers and practitioners critical insights for selecting efficient and sustainable approaches to UAV operations in complex inspection environments.

In this paper, we propose a hybrid genetic algorithm (HGA) that embeds the tabu search mechanism into the genetic algorithm (GA) for multiple-input multiple-output (MIMO) detection. We modified the selection and crossover operation to maintain the diverse and wide exploration areas, which is an advantage of the GA, and the mutation operation to perform a local search for a specific region. In the mutation process, the ’tabu’ concept is also employed to prevent the repeated search of the same area. In addition, a layered detection process is applied simultaneously with the proposed algorithm, which not only improves the bit error rate performance but also reduces the computational complexity. We apply the layered HGA (LHGA) to the MIMO system with very high modulation order such as 64-quadrature amplitude modulation (QAM), 256-QAM, and 1024-QAM. Simulation results show that the LHGA outperforms conventional detection approaches. Especially, in the 1024-QAM MIMO system, the LHGA has less than 10% of computational complexity but a 6 dB signal-to-noise ratio (SNR) gain compared to the conventional GA-based MIMO detection scheme.

The building sector is a major contributor to resource consumption, energy use, and greenhouse gas emissions. Sustainable architecture offers a solution, leveraging Building Energy Modeling (BEM) for early-stage design optimization. This study explores the use of genetic algorithms for optimizing sustainable design strategies holistically. A comprehensive analysis and optimization model was developed using genetic algorithms to individually optimize various sustainable strategies. The optimized strategies were then applied to a pre-existing building in Kabul City, a region facing significant environmental challenges. To enhance accuracy, this study integrated energy simulations with Computational Fluid Dynamics (CFD). This research combines genetic algorithms with energy simulation and CFD analysis to optimize building design for a specific climate. Furthermore, it validates the optimized strategies through a real-world case study building. Optimizing the Window-to-Wall Ratio (WWR) and shading devices based on solar exposure significantly improved the building’s energy performance. South (S)-facing single windows and specific combinations of opposing and adjacent windows emerged as optimal configurations. The strategic optimization of building component materials led to substantial energy savings: a 58.6% reduction in window energy loss, 78.3% in wall loss, and 69.5% in roof loss. Additionally, the optimized pre-existing building achieved a 48.1% reduction in cooling demand, a 97.5% reduction in heating demand, and an overall energy reduction of 84.4%. Improved natural ventilation and controlled solar gain led to a 72.2% reduction in peak-month CO2 emissions. While this study focused on applicable passive design strategies, the integration of advanced technologies like Phase Change Materials (PCMs), kinetic shading devices, and renewable energy systems can further improve building performance and contribute to achieving net-zero buildings.

This paper presents an effective hybrid metaheuristic algorithm combining the genetic algorithm (GA) and a simple algorithm based on evolutionary computation. The evolutionary approach (EA) is applied to form the initial population of the GA, thus improving the algorithm’s performance, especially its convergence speed. To assess its effectiveness, the proposed hybrid algorithm, the EAGA, is evaluated on selected benchmark functions, as well as on a real optimisation process. The EAGA is used to identify parameters in a nonlinear system of differential equations modelling an E. coli fed-batch fermentation process. The obtained results are compared against published results from hybrid metaheuristic algorithms applied to the selected optimisation problems. The EAGA hybrid outperforms the competing algorithms due to its effective initial population generation strategy. The risk of premature convergence is reduced. Better numerical outcomes are achieved. The investigations validate the potential of the proposed hybrid metaheuristic EAGA for solving real complex nonlinear optimisation tasks.

This study aims to optimize the length partitioning of modular MEP systems in building construction using a genetic algorithm, addressing challenges such as avoiding fittings (e.g., tees and crosses) and achieving standardized module lengths. To this end, this paper proposes an optimization method utilizing the customization of Revit 2021. The method comprehensively considers factors such as the location of pipe fittings, module length, production, transportation, and installation, achieving a more systematic partitioning of modules. The results show that the optimized partitioning scheme effectively avoids critical pipe fittings, and the optimized module lengths are comparable to those created manually. However, the optimized scheme includes more standardized segments, which is conducive to factory-standardized production. Additionally, cost analysis reveals that production and transportation costs account for a significant proportion of total costs, while lifting costs are relatively low. Furthermore, the presence of modules with non-standard lengths introduces corresponding penalty costs. This paper discusses the advantages and limitations of the proposed method and suggests future directions for further optimizing the algorithm and improving module partitioning. The novelty of this research lies in the integration of a genetic algorithm with BIM software to optimize MEP module partitioning, offering a more efficient and systematic approach to the modular construction process.

In this paper, the driving strategy simulations for a single-seat, lightweight, energy-efficient experimental electric vehicle are introduced. The vehicle’s operation is simulated using a developed measurement-based vehicle model in the simulation environment. The optimization was performed for the UniTrack platform at the ZalaZone proving ground using the algorithms Particle Swarm Optimization (PSO) and Genetic Algorithm (GA), with different optimization settings corresponding to varying iterations and initial population/swarm sizes. A 2.95% difference was observed between the least effective and the best PSO results, where both the number of iterations and swarm size were doubled. This demonstrates the effectiveness of PSO in solving the presented driving strategy problem, even when using fewer iterations and a smaller swarm size.

Forest fires not only cause severe damage to ecosystems and biodiversity but also directly threaten the safety of human societies. Given the significant increase in both the frequency and intensity of forest fires worldwide, especially under extreme climate conditions, efficient fire detection and initial attack (IA) are particularly critical. The initial attack is a key stage in forest fire control, and the time taken for fire detection is a crucial factor influencing the success of the initial attack. In response to the challenges of forest fire prevention and control, this study explores Unmanned Aerial Vehicle (UAV) cruising strategies, aiming to develop appropriate approaches based on regional characteristics and provide efficient periodic monitoring solutions for areas with high ecological value and challenging accessibility. By optimizing UAV patrol routes, this research seeks to maximize coverage in areas with lower initial attack success rates and significantly reduce fire detection time, thereby improving detection efficiency. We developed and applied four optimization strategies, random search, high-risk first (HRF), nearest high-risk first (NHRF), and a genetic algorithm-based (GA-based) strategy, to compare different UAV flight routes. To evaluate the deployment effectiveness of the four UAV cruise strategies, we introduced two evaluation metrics: Average Grid Risk (AGR) and Average Distance Risk (ADR). Experimental results showed that the NHRF and GA-based strategies performed better. Specifically, NHRF achieved the highest high-risk coverage, ranging from 51.5% to 71.3%, significantly outperforming the random search strategy (4–7%) and the HRF strategy (23.1–37.5%). The GA-based algorithm achieved the highest grid coverage, ranging from 30% to 59.8%, far surpassing the random search strategy (4–6.6%) and the HRF strategy (10.2–19.1%). Additionally, the NHRF and GA-based strategies delivered the best AGR and ADR performance, respectively. The application of these innovative strategies and evaluation metrics enhances forest fire prevention through periodic monitoring and supports more efficient firefighting efforts.

We present a novel approach to performing fitness approximation in genetic algorithms (GAs) using machine learning (ML) models, focusing on dynamic adaptation to the evolutionary state. We compare different methods for (1) switching between actual and approximate fitness, (2) sampling the population, and (3) weighting the samples. Experimental findings demonstrate significant improvement in evolutionary runtimes, with fitness scores that are either identical or slightly lower than those of the fully run GA—depending on the ratio of approximate-to-actual-fitness computation. Although we focus on evolutionary agents in Gymnasium (game) simulators—where fitness computation is costly—our approach is generic and can be easily applied to many different domains.

In the rapidly evolving landscape of Industry 4.0 and the transition towards Industry 5.0, manufacturing systems face the challenge of adapting to dynamic, hyper-customized demands. Current Simulation Optimization (SO) systems struggle with the flexibility needed for quick reconfiguration, often requiring time-consuming, resource-intensive efforts to develop custom models. To address this limitation, this study introduces an innovative SO design strategy that integrates three flexible simulation modeling techniques—template-based, structural modeling, and parameterization. The goal of this integrated design strategy is to enable the rapid adaptation of SO systems to diverse production environments without extensive re-engineering. The proposed SO versatility is validated across three manufacturing scenarios (flow shop, job shop, and open shop scheduling) using modified benchmark instances from Taillard’s dataset. The results demonstrate notable effectiveness in optimizing production schedules across these diverse scenarios, enhancing decision-making processes, and reducing SO development efforts. Unlike conventional SO system design, the proposed design framework ensures real-time adaptability, making it highly relevant to the dynamic requirements of Industry 5.0. This strategic integration of flexible modeling techniques supports efficient decision support, minimizes SO development time, and reinforces manufacturing resilience, therefore sustaining competitiveness in modern industrial ecosystems.

This study proposes a robust precast concrete (PC) production scheduling model for PC construction projects with schedule uncertainty by adopting a proactive scheduling approach. The proposed model consists of a PC production simulation module that simulates and evaluates the total tardiness of the schedule at a certain confidence level with contractor schedule uncertainty and a GA-based production schedule optimization module that finds an optimal schedule through iterative schedule generation and evaluation using a PC production simulation module. The experimental study shows that the proposed model can find the best schedule with 303.8 h of tardiness at a 100% confidence level, followed by NEHedd 324.7 h, basic GA 328.5 h, and EDD 335.8 h. The results of this study will help PC production schedulers to perform robustly despite contractors’ schedule changes and will thus contribute to the successful completion of PC construction projects.

To meet the preference demands of the crew members of high-speed railway trains while forming a crew team, and to automate the compilation of adaptable crew member schemes, a bilateral matching decision method for crew members is proposed based on complete preference order information. This method first describes the mutual selection process between the chief stewards and stewards of a high-speed railway train as a one-to-many bilateral matching decision process between the chief stewards and stewards Subsequently, by constructing a virtual train chief stewards, the original one-to-many bilateral matching relationship between the chief stewards and stewards is transformed into a one-to-one bilateral matching relationship between the virtual chief stewards for modeling. Then, a dual-objective integer programming model is established with the minimum sum of preference order values as the objective. Finally, an optimization solver is used to calculate the problem under different scales, and a genetic algorithm is designed for large-scale scenarios. The analysis results of numerical examples show that the model and algorithm of train crew members based on bilateral matching decisions can meet the actual requirements of the crew department and have good application value.

In this paper, a comparative analysis of the effects of groundwater, seepage and hydraulic heave on the optimal design of embedded retaining walls is carried out. The optimization model for an optimal retaining wall (ORW) minimizes the total length of the retaining wall considering design constraints. The model is extended to include the probability of failure as an additional constraint. This overcomes the limitations of the partial safety factor approach, which does not fully account for uncertainties in the soil. In contrast, the reliability-based design (RBD) approach integrates these uncertainties and enables an assessment of the impact of seepage and hydraulic heave on the reliability of the structure. A real-coded genetic algorithm was used to determine optimal designs for both optimization methods. The results of the case study show that the addition of seepage (groundwater flow) to the hydrostatic conditions has a modest effect on the embedment depth. The design based on partial safety factors, which takes seepage into account, leads to a slight increase in the embedment depth of 0.94% compared to a retaining wall design that only takes the hydrostatic conditions of the groundwater into account. When designing on the basis of probability failure, the percentage increase in embedment depth due to seepage is between 2.19% and 6.41%, depending on the target probability of failure. Furthermore, the hydraulic heave failure mechanism did not increase the required embedment depth of the retaining wall, which means that the failure mechanism of rotation near the base was decisive for the design.

The ‘black box’ nature of machine learning (ML) approaches makes it challenging to understand how most artificial intelligence (AI) models make decisions. Explainable AI (XAI) aims to provide analytical techniques to understand the behavior of ML models. XAI utilizes counterfactual explanations that indicate how variations in input features lead to different outputs. However, existing methods must also highlight the importance of features to provide more actionable explanations that would aid in the identification of key drivers behind model decisions—and, hence, more reliable interpretations—ensuring better accuracy. The method we propose utilizes feature weights obtained through adaptive feature weight genetic explanation (AFWGE) with the Pearson correlation coefficient (PCC) to determine the most crucial group of features. The proposed method was tested on four real datasets with nine different classifiers for evaluation against a nonweighted counterfactual explanation method (CERTIFAI) and the original feature values’ correlation. The results show significant enhancements in accuracy, precision, recall, and F1 score for most datasets and classifiers; this indicates the superiority of the feature weights selected via AFWGE with the PCC over CERTIFAI and the original data values in determining the most important group of features. Focusing on important feature groups elaborates the behavior of AI models and enhances decision making, resulting in more reliable AI systems.

The wind farm layout optimization problem (WFLOP) aims to maximize wind energy utilization efficiency under different wind conditions by optimizing the spatial layout of wind turbines to fully mitigate energy losses caused by wake effects. Some high-performance continuous optimization methods, such as differential evolution (DE) variants, exhibit limited performance when directly applied due to WFLOP’s discrete nature. Therefore, metaheuristic algorithms with inherent discrete characteristics like genetic algorithms (GAs) and particle swarm optimization (PSO) have been extensively developed into current state-of-the-art WFLOP optimizers. In this paper, we propose a novel DE optimizer based on a genetic learning-guided competitive elimination mechanism called CEDE. By designing specialized genetic learning and competitive elimination mechanisms, we effectively address the issue of DE variants failing in the WFLOP due to a lack of discrete optimization characteristics. This method retains the adaptive parameter adjustment capability of advanced DE variants and actively enhances population diversity during convergence through the proposed mechanism, preventing premature convergence caused by non-adaptiveness. Experimental results show that under 10 complex wind field conditions, CEDE significantly outperforms six state-of-the-art WFLOP optimizers, improving the upper limit of power generation efficiency while demonstrating robustness and effectiveness. Additionally, our experiments introduce more realistic wind condition data to enhance WFLOP modeling.

Path planning technology is of great consequence in the field of unmanned aerial vehicles (UAVs). In order to enhance the safety, path smoothness, and shortest path acquisition of UAVs undertaking tasks in complex urban multi-obstacle environments, this paper proposes an innovative composite improvement algorithm that integrates the advantages of the jellyfish search algorithm and the particle swarm algorithm. The algorithm effectively overcomes the shortcomings of a single algorithm, including parameter setting issues, slow convergence rates, and a tendency to become trapped in local optima. Additionally, it enhances the path smoothness, which improves the path optimisation. This enhances the capacity of UAVs to optimise their paths in environments characterised by multiple obstacles. To evaluate the practical effectiveness of the algorithm, a three-dimensional complex city model was constructed for the purposes of the study, and an adaptation function was designed for the purpose of evaluation. The experimental evaluation of 23 benchmark functions, the simulation test of the 3D city model, and 100 repetitive experiments demonstrate that the composite improved algorithm has a considerable advantage over the other comparative algorithms regarding performance. It exhibits fast convergence, high accuracy, and both global and local search capabilities, which enable the effective planning of a UAV flight path and the maintenance of good stability. In comparison to traditional algorithms, the composite improved algorithm demonstrably reduces the flight time and the number of obstacle avoidance manoeuvres required by the UAV. It provides robust technical support for the path planning of the UAV in complex urban environments and facilitates the advancement and implementation of related technologies.

The flower-growing sector in Latin America presents significant health risks for workers, which highlights the need for technological updates in their production processes. Likewise, outdated machinery leads to losses that need to be avoided. The method of productive innovation developed in this document involves optimizing a mechanism of agricultural machinery used in carnation classification. The optimization is achieved by minimizing the jerk of the mechanism’s movement using metaheuristic methods. The results of three metaheuristic methods are compared against a brute force methodology. Optimization using these metaheuristic methods allows for achieving satisfactory results with up to 98% time reduction in the optimization process. This jerk optimization gives a longer useful life to the machinery, reduces the production stops needed for maintenance from once an hour to once every three hours, and reduces the damage done by the machine to the carnation stems.

Autonomous driving technology, as a key component of intelligent transportation systems, has gained considerable attention in recent years. While significant progress has been made in areas such as path planning, obstacle detection, and navigation, relatively less focus has been placed on vehicle speed control, which plays a critical role in ensuring safe and efficient operation, especially in complex and dynamic road environments. This paper addresses the challenge of speed tracking by proposing a genetic algorithm-based optimization method for PID controller parameters. Traditional PID controllers often struggle with maintaining accuracy and response time in highly variable conditions, but by optimizing these parameters through the genetic algorithm, substantial improvements in speed control precision and adaptability can be achieved, enhancing the vehicle’s ability to navigate real-world driving scenarios with greater stability. The experimental results clearly indicate that the autonomous vehicle, after PID parameter optimization using a genetic algorithm, demonstrated the following speed tracking errors: when the road surface adhesion coefficient was 0.5, the maximum speed tracking error was 0.22, the average value was 0.063, and the standard deviation was 0.124; when the adhesion coefficient was 0.6, the maximum speed tracking error was 0.180, the average value was 0.056, and the standard deviation was 0.099; when the adhesion coefficient was 0.8, the maximum speed tracking error was 0.179, the average value was 0.056, and the standard deviation was 0.098. This method significantly improved the controller’s performance in maintaining the desired speed, even under challenging conditions. These findings highlight the potential of genetic algorithms in supporting the future development of autonomous driving technology, ensuring its successful integration into intelligent transportation systems.

This paper presents a fault diagnosis technique for milling machines based on acoustic emission (AE) signals and a hybrid deep learning model optimized with a genetic algorithm. Mechanical failures in milling machines, particularly in critical components like cutting tools, gears, and bearings, account for a significant portion of operational breakdowns, leading to unplanned downtime and financial losses. To address this issue, the proposed method first acquires AE signals from the milling machine. AE signals, capturing the dynamic responses of machine components, are transformed into continuous wavelet transform (CWT) scalograms for further analysis. Gaussian filtering is applied to enhance the clarity of these scalograms, effectively reducing noise while maintaining essential features. A convolutional neural network (CNN) based on the VGG16 architecture is utilized for spatial feature extraction, followed by a bidirectional long short-term memory (BiLSTM) network to capture the temporal dependencies of the scalograms. The genetic algorithm (GA) is used to optimize feature selection and ensure the selection of the most relevant features to further improve the model’s performance. The optimized features are finally fed into a fully connected (FC) layer of the proposed hybrid model for fault classification. The proposed method achieves an accuracy of 99.6%, significantly outperforming traditional approaches. This method offers a highly accurate and efficient solution for fault detection in milling machines, allowing for more reliable predictive maintenance and operational efficiency in industrial settings.

The evolving landscape of cyber threats necessitates continuous advancements in defensive strategies. This paper explores the potential of artificial intelligence (AI) as an emerging tool to enhance cybersecurity. While AI holds widespread applications across information technology, its integration within cybersecurity remains a recent development. We offer a comprehensive review of current AI applications in this domain, focusing particularly on their preventative capabilities against prevalent threats like phishing, social engineering, ransomware, and malware. To illustrate these concepts, the paper presents a case study showcasing a specific AI application in a cybersecurity context. This case study addresses a critical gap in securing communication within resource-constrained Internet of Things (IoT) networks using the IEEE 802.15.4 standard. We discussed the advantages and limitations of employing PN sequence encryption for this purpose.

Aimed at the high cost of cold chain distribution of fresh agricultural products within a specified time window, a joint optimization method based on a bi-level programming model for cold chain logistics is proposed for the location of front warehouses and distribution path planning. At the upper level of the bi-level programming model, k-means clustering analysis is used to obtain all accurate information about alternative locations for the front warehouse for site selection, thereby providing the corresponding foundation for the lower level algorithm. At the lower level of the model, a fusion algorithm of particle swarm optimization (PSO) and a genetic algorithm (GA) is used for solving. To accelerate the convergence speed of the population and lower the running time of the algorithm, the parameter values in the algorithm are determined adaptively. An adaptive hybrid algorithm combining the particle swarm optimization algorithm and the genetic algorithm (APSOGA) is used to reallocate the location information on backup points for the front-end warehouse, ultimately determining the facility location of the front-end warehouse and planning the end path from the front-end warehouse to the customer point, achieving joint optimization of the front-end warehouse’s location and path. A comparative analysis of algorithm optimization shows that using the APSOGA hybrid algorithm can reduce the total cost of the logistics network by 14.57% compared to a traditional single-algorithm PSO solution and reduce it by 5.21% compared to using a single GA. This proves the effectiveness of the APSOGA hybrid algorithm in solving location and path planning problems for cold chain logistics distribution companies.

Emotion classification is a challenge in affective computing, with applications ranging from human–computer interaction to mental health monitoring. In this study, the classification of emotional states using electroencephalography (EEG) data were investigated. Specifically, the efficacy of the combination of various feature selection methods and hyperparameter tuning of machine learning algorithms for accurate and robust emotion recognition was studied. The following feature selection methods were explored: filter (SelectKBest with analysis of variance (ANOVA) F-test), embedded (least absolute shrinkage and selection operator (LASSO) tuned using Bayesian optimization (BO)), and wrapper (genetic algorithm (GA)) methods. We also executed hyperparameter tuning of machine learning algorithms using BO. The performance of each method was assessed. Two different EEG datasets, EEG Emotion and DEAP Dataset, containing 2548 and 160 features, respectively, were evaluated using random forest (RF), logistic regression, XGBoost, and support vector machine (SVM). For both datasets, the experimented three feature selection methods consistently improved the accuracy of the models. For EEG Emotion dataset, RF with LASSO achieved the best result among all the experimented methods increasing the accuracy from 98.78% to 99.39%. In the DEAP dataset experiment, XGBoost with GA showed the best result, increasing the accuracy by 1.59% and 2.84% for valence and arousal. We also show that these results are superior to those by the previous other methods in the literature.

Global optimization is critical in engineering, computer science, and various industrial applications as it aims to find optimal solutions for complex problems. The development of efficient algorithms has emerged from the need for optimization, with each algorithm offering specific advantages and disadvantages. An effective approach to solving complex problems is the hybrid method, which combines established global optimization algorithms. This paper presents a hybrid global optimization method, which produces trial solutions for an objective problem utilizing a genetic algorithm’s genetic operators and solutions obtained through a linear search process. Then, the generated solutions are used to form new test solutions, by applying differential evolution techniques. These operations are based on samples derived either from internal line searches or genetically modified samples in specific subsets of Euclidean space. Additionally, other relevant approaches are explored to enhance the method’s efficiency. The new method was applied on a wide series of benchmark problems from recent studies and comparison was made against other established methods of global optimization.

This review examines the increasing application of artificial intelligence (AI) and/or machine learning (ML) in microalgae processes, focusing on their ability to improve production efficiency, yield, and process control. AI/ML technologies are used in various aspects of microalgae processes, such as real-time monitoring, species identification, the optimization of growth conditions, harvesting, and the purification of bioproducts. Commonly employed ML algorithms, including the support vector machine (SVM), genetic algorithm (GA), decision tree (DT), random forest (RF), artificial neural network (ANN), and deep learning (DL), each have unique strengths but also present challenges, such as computational demands, overfitting, and transparency. Despite these hurdles, AI/ML technologies have shown significant improvements in system performance, scalability, and resource efficiency, as well as in cutting costs, minimizing downtime, and reducing environmental impact. However, broader implementations face obstacles, including data availability, model complexity, scalability issues, cybersecurity threats, and regulatory challenges. To address these issues, solutions, such as the use of simulation-based data, modular system designs, and adaptive learning models, have been proposed. This review contributes to the literature by offering a thorough analysis of the practical applications, obstacles, and benefits of AI/ML in microalgae processes, offering critical insights into this fast-evolving field.

Aiming to address the problem of selecting the parameters of a soil constitutive model in the calculation of foundation pit stability, this paper proposes a hybrid genetic differential evolution algorithm (GADE) which performs by “jumping out of local optima” with “fast convergence” based on the hybrid optimization algorithm strategy and compares the advantages and disadvantages of genetic algorithms (GAs) and differential evolution algorithms (DEs). Three typical test functions were used to evaluate the search efficiency and convergence speed of GAs, DEs, and GADE, respectively. It was found that GADE has the fastest convergence speed and can search for the global optimal solution to the problem, which highlights its excellent optimization performance. At the same time, taking the Shimao Binjiang deep foundation pit as an example, GADE was used to invert the soil modulus parameters of a CX1 measuring point and construct a finite-element model for calculation. The results showed that the simulated calculation curve and the measured displacement curve were in good agreement and the curve fitting reached 95.05%, indicating the applicability and feasibility of applying GADE to identify soil parameters.

The increasing interests and recent advancements in artificial intelligence and machine learning have significantly accelerated the development of novel techniques for the state estimation of batteries in electrified vehicles’ battery management systems (BMSs). Determining the remaining capacity among the several BMS states is crucial for ensuring the safe and stable functioning of an electric vehicle. This paper proposes an adaptive estimator for the remaining capacity of lithium-ion batteries, leveraging a Genetic Algorithm (GA)-tuned random forest (RF) regressor. The estimator is designed to function effectively under varying thermal conditions. The optimization of critical parameters, namely, the number of estimators (n-estimators) and the minimum number of samples per leaf (min-samples-leaf), is a focal point of this study to enhance model accuracy and robustness. The model effectively captures the battery’s dynamic behavior and inherent non-linearity. The Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) achieved during testing demonstrate promising accuracy and superior prediction. The results demonstrated significant improvements in state of charge (SOC) estimation accuracy. The proposed GA-optimized RF model achieved an MAE of 0.0026 at 25 °C and 0.0102 at −20 °C, showing a 41.37% to 50% reduction in the MAE compared to traditional random forest models without GA optimization. The RMSE was also reduced by 18.57% to 31.01% across the tested temperature range. These improvements highlight the model’s ability to accurately estimate the SOC in varying thermal conditions.

This paper introduces a hybrid optimization method that leverages either linear programming (LP) or a genetic algorithm (GA) based on the problem size to enhance the parallel additive manufacturing (AM) process for ship models. The LP ensures optimality but can experience exponential increases in the computation time as the problem size grows. To address this limitation, the GA is employed for larger problems, providing optimal solutions within reasonable quality and time constraints. The method optimizes the module allocation to AM machines and determines the build processing sequence for each machine, while also considering the availability of workers preparing for consecutive module production. Applied to a case study, the proposed method achieves a 14% reduction in the completion time compared to a heuristic method from a previous study. Furthermore, the method is validated by benchmarking against the heuristic method across various problem sizes, consistently demonstrating superior performance.