Multi-objective Genetic Algorithm Optimization Research Articles

Thermodynamic calculation-assisted design of 500 MPa high performance steel by machine learning

In order to rationally design Q500 low-alloy wind power steel with stable microstructure and properties within a large cooling rate range, this study proposed a design system that utilizes thermodynamic database, machine learning (ML), and the multi-objective genetic optimization algorithm (MOGA). Thermodynamic calculation has been used to obtain a large amount of data on phase fractions and mechanical properties for various chemical composition at different cooling rates. By comparing the predictive capabilities of different ML models, the deep neural network (DNN) and MOGA were combined to design Q500 wind power steel. Subsequently, three groups of designed chemical compositions were appropriately selected for verification. The results indicate that the yield strength of the experimental steel was over 500 MPa, and the maximum error in microstructure and mechanical property prediction was 5.0%. The prediction results demonstrated the rationality of the composition design framework. Meanwhile, the method of combining the material database and ML proposed in this study can also be applied to the design of other low-alloy steels.

AI-Driven Optimization of Roll Forming Parameters for Defect-Free Manufacturing in Advanced High-Strength Steels

This paper presents an innovative and novel AI-driven approach to optimizing the roll forming process for advanced high-strength steels (AHSS), which are prone to defects like wrinkling, springback, and cracking due to their high strength and low formability. Traditional trial-and-error methods for parameter selection are time-consuming and inefficient. The proposed framework integrates machine learning models, such as artificial neural networks (ANN), support vector machines (SVM), and random forests, with genetic algorithms for multi-objective optimization. The novelty of this paper lies in the real-time monitoring and adjustment capabilities, allowing for dynamic process control that adapts to variations in material properties. This approach not only predicts and reduces defects by 30%, but also decreases parameter selection time by 50%, significantly improving production efficiency. A case study on an AHSS automotive component demonstrates the effectiveness of this system in achieving defect-free manufacturing, reducing equipment costs, and enhancing throughput. The AI-based methodology represents a paradigm shift in roll forming, offering a more flexible, scalable, and cost-efficient solution for modern manufacturing. Future work will explore integration with finite element analysis (FEA) and reinforcement learning for further improvements in process control.

Optimization study on circadian tunability and Rec. 2020 of RGB-LED-based displays considering the users with different ages

The reasonable design of spectral power distribution (SPD) of GaN-based light-emitting diode (LED) display becomes more and more important and popular. In this work, we perform an investigation on the circadian effects of three-primary (red R, green G, and blue B, also denoted as RGB) LED based self-luminous displays (such as emerging mini-LED or micro-LED based displays) considering the users with different ages. With the increment of peoples’ ages from 1 year to 100 years, the optical transmittance of humans’ eyes may decrease due to the aging of their body functions. Therefore, it is important and meaningful for performing a spectral optimization study on the circadian effects of light that is emitting from self-luminous LED-based displays for the users with different ages. Here, the melanopic efficacy of luminous radiation (MELR) is mainly used to evaluate the circadian effects of white light coming from RGB-LED displays. With the aid of multi-objective optimization genetic algorithm (MOGA), the maximum MELR and the minimum MELR value are separately obtained for the standard observers at the age of 32, while maintaining at a certain high color gamut of Rec. 2020 standard (i.e., 60 %, 70 %, 80 %, and 90 % Rec. 2020, respectively) for these LED-based displays. Then, an extrapolation of optimal SPD at the age of 32 is done to other ages. Another parameter that can be denoted as circadian stimulus (CS) is also calculated and discussed for different ages. It is believable that this work is helpful for guiding the design and fabrication of future RGB-LED-based displays in consideration of the users with different ages, meeting large requirements of human-centric lighting (HCL) in the future.

Enhancing pre-trained models for text summarization: a multi-objective genetic algorithm optimization approach

Enhancing pre-trained models for text summarization: a multi-objective genetic algorithm optimization approach

Multi-objective parameter design and economic analysis of VSG-controlled hybrid energy systems in islanded grids

Photovoltaic (PV) systems offer cost-effective power solutions for outlying islands but often compromise system stability due to reduced inertia. This study introduces a Virtual Synchronous Generator (VSG) control strategy, integrated with Energy Storage Systems (ESS) and PV, to enhance system inertia. By optimizing coordination between these energy sources, the proposed method mitigates oscillations and improves grid stability. However, PV-VSG systems are generally not favored by energy providers due to the requirement for pre-curtailment of power output. To address this, the paper proposes a parameter design method for VSG control of ESS and PV, utilizing multi-objective genetic algorithm (MOGA) optimization to simultaneously increase the frequency nadir and minimize the settling time after disturbances. Additionally, an adaptive curtailment decision and parameter design method based on artificial neural networks is introduced to enhance the feasibility of PV-VSG systems by reducing PV pre-curtailment and prioritizing PV power release and ESS charging during frequency oscillations. Real data from the Penghu Archipelago in Taiwan are used to build a dynamic model in DIgSILENT, enabling interaction with MOGA. The Value at Risk (VaR) method with dual stochastic variables is employed to assess the allowable PV installed capacity. The results show that when VaR is set at 1%, the proposed PV-VSG method can increase PV penetration by 57.5% compared to scenarios without VSG. Furthermore, compared to traditional PV-VSG methods, the proposed approach achieves a 16.8% increase in PV penetration and reduces annual PV curtailment by 25 MWh. This study also evaluates the economic impact of planners choosing different risk levels, offering valuable insights for grid development in remote or island regions.

Optimizing thermal energy storage in rectangular unit: Exploring the impact of moving fin with phase change material

This research presents an innovative moving fin design for a latent heat storage unit, with the objective of improving thermal performance. The methodology integrates numerical simulation, a machine learning approach, and a multi-objective genetic optimization algorithm (NSGA-II). The study optimizes fin parameters, such as fin velocity (Vf), fin thickness (tf), fin length (Lf) and initial fin position (yf). The optimization process aims to maximize both power (P) and stored energy per mass (Em), assigning equal importance to both criteria. The outcomes reveal a considerable rise in power with minor compromises in stored energy per mass. The incorporation of moving fins in rectangular enclosure with PCM leads to a significant power boost while preserving stored latent energy with minimal compromise. In comparison to the no-fin condition, fixed 1-fin and fixed 3-fin configurations lead to stored energy losses of 4.16 % and 11.73 %, respectively, accompanied by power increases of 20.48 % and 51.17 %, respectively. Conversely, the optimized condition for maximum power with moving fins incurs a mere 2.73 % stored energy loss but achieves a remarkable 173.92 % increase in power. The optimized design parameters for maximizing power (P) include Vf = 22.5 × 10−3 mm/s, yf = 60 mm, tf = 2 mm, and Lf = 40 mm.

Advanced energy management system with road gradient consideration for fuel cell hybrid electric vehicles

Hybrid electric fuel cell vehicles (FCHEVs) represent a promising pathway for sustainable transportation by integrating fuel cells, batteries, and supercapacitors to power their propulsion systems. However, ensuring the efficiency and longevity of FCHEV components remains a significant challenge due to stress on batteries and fuel cells, as well as fluctuations in the DC bus voltage. This paper presents an innovative approach to enhance the effectiveness and lifespan of FCHEV storage elements through an energy management system (EMS) based on a multi-objective genetic optimization algorithm that accounts for road gradient variations. The proposed methodology dynamically allocates energy among the fuel cell, battery, and supercapacitor, optimizing three key objectives simultaneously: minimizing fluctuations in fuel cell and battery power, reducing variations in DC bus voltage, and balancing performance, efficiency, and storage system longevity. Extensive simulations demonstrate that simultaneous optimization of these three objectives yields the best results. Under the NEDC cycle, the optimized EMS resulted in a 12 % increase in fuel cell stress but a significant 73 % decrease in battery stress, along with a 39 % reduction in DC bus voltage variations. For the UDDS cycle, there was a 37 % decrease in fuel cell stress, a 69 % decrease in battery stress, and a 29 % reduction in DC bus voltage variations. During the HWFET cycle, the optimized EMS achieved an 8 % decrease in fuel cell stress, a 49 % decrease in battery stress, and an 8 % reduction in DC bus voltage variations. By incorporating road gradient considerations, the EMS aims to maintain the battery state of charge and hydrogen consumption within acceptable ranges while enhancing FCHEV efficiency under various road gradient conditions. This study advances FCHEV technology, offering insights into the design and operation of more sustainable and efficient transportation solutions.

Probabilistic short-term load forecasting models based on a parametric approach

The forecasting of electric load plays an essential role in the effective management of electric power systems. Specifically, short-term models, which predict hourly load over the 24 hours of the subsequent day, hold significant value for applications within the realm of electricity markets. In this context, research efforts have predominantly concentrated on the development of point load forecasting models. These models provide solely the estimated value of hourly load, omitting any information regarding its associated uncertainty. Probabilistic forecasting models aim to address this limitation by offering comprehensive information on forecasted values, including their associated uncertainty, thereby enabling their more effective utilization in risky decision-making environments. This paper presents a parametric probabilistic model designed for hourly load forecasting. The model is refined through a multi-objective genetic algorithm optimization process that identifies explanatory variables from a specified set. The selected variables are combined linearly to predict the parameters of a probability distribution function for the hourly load. The process also selects the type of distribution from among those characterized by two parameters. The model is applied to data from a real distribution substation, yielding superior forecasting evaluation indexes compared to those achieved by two benchmark probabilistic load forecasting models.

Approach for sizing by optimization of an electrical drive considering a multiphysics and multidynamics behaviour

PurposeThis paper aims to describe a modelling and solving methodology of a (static converter–electric motor–control) system for its sizing by optimization, considering the dynamic thermal heating of the machine.Design/methodology/approachThe electrical drive sizing model is composed of two simulators (electrical and thermal) that are co-simulated with a master−slave relationship for the time step management. The computation is stopped according to simulation criteria.FindingsThis paper details a methodology to represent and size an electrical drive using a multiphysics and multidynamics approach. The thermal simulator is the master and calls the electrical system simulator at a fixed exchange time step. The two simulators use a dedicated dynamic time solver with adaptive time step and event management. The simulation automatically stops on pre-established criteria, avoiding useless simulations.Research limitations/implicationsThis paper aims to present a generic methodology for the sizing by optimization of electrical drives with a multiphysics approach, so the precision and computation time highly depend on the modelling method of each components. A genetic multiobjective optimization algorithm is used.Practical implicationsThe methodology can be applied to size electrical drives operating in a thermally limited zone. The power electronics converter and electrical machine can be easily adapted by modifying their sub-model, without impacting the global model and simulation principle.Originality/valueThe approach enables to compute a maximum operating duration before reaching thermal limits and to use it as an optimization constraint. These system considerations allow to over constrain the electrical machine, enabling to size a smaller machine while guaranteeing the same output performances.

Multiobjective optimization of porous medium for efficient transpiration cooling of hypersonic vehicles using genetic algorithm

Transpiration cooling has been proven an effective method for reducing heat flux on the surfaces of high-speed vehicles. This study investigates the effects of porous medium properties, employing a black-box optimization method to determine the optimal length, thickness, and porosity for a porous medium in a transpiration cooling system on a flat plate under hypersonic laminar flow. The objectives include optimizing thermal effectiveness, coolant consumption, and the weight and cost of the porous material. A multiobjective genetic optimization algorithm is directly integrated with a Computational Fluid Dynamics (CFD) solver, and 1562 CFD simulations were conducted to identify the optimal configuration. The results demonstrate that the length and porosity of the porous medium more significantly impact thermal effectiveness than the thickness. Furthermore, the optimization identified a configuration for the porous medium that, when compared to the original case, shows reductions in length, thickness, and porosity of 3.5%, 11%, and 29%, respectively. Additionally, there were average improvements in thermal effectiveness and coolant consumption of 4.56% and 3.9%, respectively, while the weight and cost of the porous material increased by 3.73% and 3.65%, respectively.

An Efficient Real-Time Vehicle Classification from a Complex Image Dataset Using eXtreme Gradient Boosting and the Multi-Objective Genetic Algorithm

Recent advancements in image processing and machine-learning technologies have significantly improved vehicle monitoring and identification in road transportation systems. Vehicle classification (VC) is essential for effective monitoring and identification within large datasets. Detecting and classifying vehicles from surveillance videos into various categories is a complex challenge in current information acquisition and self-processing technology. In this paper, we implement a dual-phase procedure for vehicle selection by merging eXtreme Gradient Boosting (XGBoost) and the Multi-Objective Optimization Genetic Algorithm (Mob-GA) for VC in vehicle image datasets. In the initial phase, vehicle images are aligned using XGBoost to effectively eliminate insignificant images. In the final phase, the hybrid form of XGBoost and Mob-GA provides optimal vehicle classification with a pioneering attribute-selection technique applied by a prominent classifier on 10 publicly accessible vehicle datasets. Extensive experiments on publicly available large vehicle datasets have been conducted to demonstrate and compare the proposed approach. The experimental analysis was carried out using a myRIO FPGA board and HUSKY Lens for real-time measurements, achieving a faster execution time of 0.16 ns. The investigation results show that this hybrid algorithm offers improved evaluation measures compared to using XGBoost and Mob-GA individually for vehicle classification.

Multi-objective optimization of tribological properties of camshaft bearing pairs using DNN coupled with NSGA-II algorithm and TOPSIS

PurposeThis study aims to propose an optimization framework using deep neural networks (DNN) coupled with nondominated sorting genetic algorithm II and technique for order preference by similarity to an ideal solution method to improve the tribological properties of camshaft bearing pairs of internal combustion engine.Design/methodology/approachA lubrication model based on the theory of elastohydrodynamic lubrication and flexible multibody dynamics was developed for a V6 diesel engine. Setting DNN model as fitness function, the multi-objective optimization genetic algorithm and decision-making method were used to optimize the bearing pair structure with the goal of minimizing the total friction loss and the difference of the average values of minimum oil film thickness.FindingsThe results show that the lubrication state corresponding to the optimized bearing pair structure is elastohydrodynamic lubrication. Compared with the original structure, the optimized structure significantly reduces the total friction loss.Originality/valueThe optimized performance and corresponding structural parameters are obtained, and the optimization results were verified through multibody dynamics simulation.Peer reviewThe peer review history for this article is available at: https://publons.com/publon/10.1108/ILT-12-2023-0417/

Response Surface Modeling and Multi-objective Genetic Algorithm Optimization of Polyphenol/terpineol Yield from Luffa cylindrica Oil

Response Surface Modeling and Multi-objective Genetic Algorithm Optimization of Polyphenol/terpineol Yield from Luffa cylindrica Oil

Application of multi-objective optimization genetic algorithm to design terahertz metamaterials with fano resonances

In the prosperous development of terahertz (THz) metamaterials, Fano resonances have gained attention due to their potential applications in ultrasensitive systems. The performance of Fano resonance is directly influenced by the geometrical parameters of the element structure. However, the traditional design rules for Fano resonances in metamaterials rely on an empirical trial-and-error strategy, necessitating significant effort to achieve optimal results. To address this issue, we propose a design method in this study that utilizes the finite integration technique in time domain (FITD) along with a multi-objective optimization genetic algorithm for the intelligent design of metamaterial structures exhibiting the Fano resonance phenomenon. The FITD method is primarily used to calculate the Fano resonance with different metamaterial geometric structure parameters, while the genetic algorithm efficiently selects the optimal solution. Our method, characterized by high efficiency and complete independence from prior knowledge, could offer a new design technique for metamaterials with specific functions, thereby contributing to the development of THz applications.

The Fan Design Optimization for Totally Enclosed Type Induction Motor with Experimentally Verified CFD-Based MOGA Simulations

The energy efficient electric motors save energy thus reduce operating costs. Since there are several losses affecting the motor efficiency, fan design plays an important role to minimize these losses. This study examines the effects of the design parameters of a radial bladed fan on the motor efficiency in a 132 frame with a power of 7.5 kW. The parametric analysis was carried out with the computational fluid dynamics method, and the results were used for the multiobjective genetic algorithm (MOGA) optimization study with the highest efficiency and the lowest motor body temperature objectives. The hub height, hub radius, distance to body cover, blade rising angle, fan cover entrance distance, blade edge angle, blade center radius, blade edge radius, blade end radius, and center edge distance were selected as optimization parameters. 151 simulations were performed. The results showed that the most important parameter for fan efficiency is the hub height which is the parameter that determines the height of the fan impeller diameter. According to the results, the optimum fan design increased the efficiency by 8% compared to the original fan and reduced the winding temperature by 8 °C. The optimized fan design was manufactured and tested against imulation data. This study contributes to sustainable development goals by improving motor efficiency that reduces the cost, designing of new components, and cooling the fan effectively that reduces the amount of copper used.

Optimization of Elbow Draft Tubes for Variable Speed Propeller Turbine

The design of the elbow draft tubes is challenging due to the complexity of the flow. The whole turbine unit’s power output strongly depends on the draft tube function, especially for the low-head turbines. The article presents a novel approach to optimizing elbow draft tubes for a variable-speed propeller turbine designed for low-head applications. First, the study addresses the specifics of the propeller variable speed turbine by comparing the classical Kaplan turbine. Then, the grid scaling test is conducted to evaluate the uncertainty of the pressure regeneration. Further, a new approach to parameterising the elbow draft tube geometry is introduced. The study employs ANSYS CFX 2021 R1 software for numerical simulation to optimise the elbow draft tube geometry in the CAESES environment. After the sensitivity test and deselecting the non-sensitive parameters, we perform multi-objective genetic algorithm (MOGA) optimization. The optimization process results in a Pareto front of optimised elbow draft tube shapes with the best pressure regeneration for different draft tube construction heights, enabling the selection of suitable candidates for various locations. Minimal difference in the performance of the selected elbow draft tube shapes with the simple straight draft tube confirms a high-quality draft tube optimization achievement.

Close-range maneuver planning for uncontrolled rendezvous with multiple elliptical orbit targets based on genetic algorithm

On-orbit debris removal is an important and challenging problem in space engineering. A low-cost method for this problem is to use uncontrolled rendezvous with a sub-spacecraft released by a service spacecraft. The sub-spacecraft do not carry any control system and are designed to rendezvous with the target debris to remove it. This paper presents a comprehensive study on the mission planning. First, we analyze and dynamically model the uncontrolled rendezvous process for elliptical orbit targets, and study various aspects such as maneuver, sub-spacecraft release parameter solution, and rendezvous accuracy estimation. Then, we use NSGA-II, a multi-objective optimization genetic algorithm, to establish the mission planning model, focusing on the design of decision variables, constraints and fitness functions. Finally, we apply our proposed method to two specific cases of Molniya orbit, and verify the effectiveness of the planning model. We also conduct a deep analysis of the planning results, and summarize the commonalities and differences of various effective strategies.

Multi-Objective Optimization of a Multi-Cavity, Significant Wall Thickness Difference Extrusion Profile Mold Design for New Energy Vehicles.

With the rapid development of the new energy vehicle market, the demand for extruded profiles for battery trays, mainly characterized by significant wall thickness differences in multiple chambers, is increasing, posing new challenges to production and quality control. This study examines the multi-objective optimization problem in the design process of aluminum profile dies with multi-cavity profiles and significant wall thickness differences. Using QFORM-extrusion professional aluminum extrusion finite element analysis software and the response surface analysis method, the standard deviation of the velocity (SDV), standard deviation of the pressure (SDP), and thick wall hydrostatic pressure (TWHP) on the profile section at the die exit are optimized. By analyzing the functional relationship between the key die structure parameters (the height of the baffle plates, the length of the bearing, and the height of the false mandrel) and the optimization objective, the optimal combination scheme of die structure parameters was obtained using the NSGA2 (non-dominated sorting genetic algorithm-2) multi-objective genetic optimization algorithm. The results show that, compared with the initial design scheme, the standard deviation of profile section velocity was reduced by 5.33%, the standard deviation of pressure was reduced by 11.16%, and the thick wall hydrostatic pressure was increased by 26.47%. The die designed and manufactured using this scheme successfully completed the hot extrusion production task, and the profile quality met the predetermined requirements, thus verifying the effectiveness of this study in optimizing the design of a multi-cavity aluminum profile die with significant differences in wall thickness for complex structures.

Enhanced reaction extraction distillation via multi-objective optimization and 4E analysis for sustainable recovery of organic compounds from wastewater

Resource utilization and sustainable treatment of wastewater, which can achieve efficient energy utilization and environmental protection, have received hot social attention. Tetrahydrofuran and methanol are the two most common substances in pharmaceutical wastewater. It is very necessary to explore the wastewater treatment process containing these two substances. In this study, extractive distillation and reactive distillation were proposed to effectively separate the tetrahydrofuran-methanol-water component of pharmaceutical wastewater. To further reduce energy consumption, two strengthening processes, double-column reactive-extractive distillation and reactive-extractive dividing wall column distillation were designed. Taking the minimum operating and equipment cost as the objective function, the processes were optimized by using the multi-objective genetic optimization algorithm to obtain the best process parameters. Four processes were evaluated in terms of total annual cost, energy consumption, gas emissions and thermodynamic efficiency. Compared to the traditional three-column extractive distillation process, the above-mentioned four aspects of the reactive-extractive dividing wall column process was reduced by 40.63%, 43.13%, 43.14% and 17.71%, respectively. The reactive-extractive dividing wall column distillation process is of great significance for the efficient purification of valuable organic matter in wastewater and environmental protection.

Structural Optimization and Process Scheme of Large-span Gantry Crossbeam based on Response Surface

A comprehensive design method for large-span gantry beams, combining process techniques and structural optimization, is proposed to address significant deflection deformation issues caused by their own gravity and the gravity of front attachments. Selecting the beam section type, finite element static analysis is performed using ANSYS Workbench for both cast and welded gantry beams. The maximum static deformation values of the beams for both process schemes are compared to determine the process scheme for the large-span beam. Subsequently, multiple structural dimensions of the welded gantry beam are parameterized, and sensitivity analysis is conducted to determine the degree of influence of each parameter on the optimization objective. Parameters with relatively small effects on the corresponding objectives are eliminated. A response surface multi-objective genetic algorithm optimization based on Box-Behnken design experimental design method is employed to minimize deflection deformation and reduce weight. The response surface model is fitted with expressions, and variance and significance analyses are conducted to validate the model's reliability. The optimal Pareto solutions are determined based on the response surface model using a multi-objective genetic algorithm. A comparison between the beams before and after optimization shows a reduction of 0.052 in maximum static deformation of the welded beam, representing an 8.3% decrease. Additionally, the mass decreases by 1710 kg, indicating a 14.63% reduction, demonstrating significant optimization effects.