Multi-objective Performance Optimization Research Articles

Influence and optimization of dual-orifice jet nozzles on oil capture performance of under-race lubrication with a radial oil scoop for high-speed bearings in aero-engine

The objective of this research is to determine the influence of dual-orifice jet nozzles on the oil–air flow dynamics and oil capture performance in the under-race lubrication for high-speed bearings and to identify the optimal design parameter combinations for optimizing the oil capture performance. An experimental setup for under-race lubrication is specifically designed, complemented by a multiphase computational fluid dynamics model to elucidate the oil–air flow behavior. The response surface methodology is combined with a multi-objective particle swarm optimization algorithm to perform multi-objective optimization of the oil capture performance. The findings indicate that oil capture efficiency initially rises to a peak and subsequently declines with increasing orifice spacing, identifying an optimal orifice spacing that maximizes efficiency. While the oil flow rate from dual-orifice jet nozzles is lower than twice that of single-orifice jet nozzles, appropriate parameter matching allowed for an increase in both captured lubricant oil and efficiency. Optimized configurations achieved substantial enhancements in oil capture efficiency, with the greatest improvement exceeding 16%, all while maintaining precise oil supply to the high-speed bearings.

Design and Multi-Objective Performance Optimization of a Novel Steering Technology for Heavy Goods Vehicles

Abstract A multi-objective optimization (MOO) approach is utilized to design a controller for a novel rear-steering technology named “Brake-Actuated Steering” (BAS). This system uses individually controlled brakes to generate differential longitudinal forces on each side of an axle, causing it to steer. Compared to other active rear-steering solutions utilizing path-following control, the BAS system is expected to provide comparable maneuverability performance, while offering approximately a 50% reduction in both mass and costs. Two objective criteria that define the performance and control effort of the BAS system are considered. Constraints are imposed limiting the feasible set of design variables to ensure stability of the controller, and sufficient centering capability of the steering system in emergency braking conditions. The optimization is performed for low-speed cornering of a tractor-semitrailer under various operating conditions, including low-friction surfaces, different axle loadings, and vehicle speeds. The optimization provides a set of Pareto-optimal fronts, minimizing the objectives. Simulations are used to compare the performance of a nonoptimal design for the BAS axle prototype to that of the optimized axle design. These validate the superior performance resulting from the optimization, with root mean square error of the steering angle and the energy consumed by the towing unit reduced by 48% and 21%, respectively. Model and controller validation and the performance of the system are verified by experiments on a prototype vehicle system.

Multi-Objective Performance Optimization of Machine Learning Models in Healthcare.

Multi-objective optimization holds particular significance for medical applications, wherein enhancing sensitivity is crucial to avoid costly missed diagnoses, and maintaining high specificity is imperative to prevent unnecessary procedures. In particular, when optimizing machine learning architectures for clinical diagnostics, it becomes essential to balance target quality measures such as accuracy, sensitivity, and specificity. Therefore, we developed MOOF, a multi-objective optimization framework that employs NSGA-II and TOPSIS to simultaneously optimize the model parameters of three selected ML algorithms: random forest, support vector machine, and multilayer perceptron. Finally, we evaluated the performance of the optimized MOOF models compared to gold standard methods such as multi-score grid search and single objective optimizations. Our results show that MOOF generally outperforms other approaches by inherently providing optimal solutions, representing the trade-offs between the target objectives. In conclusion, the study supports the importance of multi-objective optimization in medical informatics, with MOOF as a powerful tool for precise ML models, potentially improving patient care and clinical decision support systems.

Multi-objective optimization of heat transfer performance of H-type fin with vortex generator based on NSGA-II

To further enhance the performance of heat exchangers and bolster waste heat recovery, vortex generators (VGs) of various shapes are added to the H-type fins. This study explores the impact of VGs’ attack angles and configurations on the thermal performance of H-type finned tubes under different Reynolds (Re) numbers. Additionally, multiple dimensionless evaluation metrics are introduced to assess the flow and heat transfer characteristics with VGs, and correlations for predicting the Nusselt (Nu) number and pressure drop at varying attack angles are developed. The results reveal that at low Re numbers, trapezoidal winglets yield the highest comprehensive performance factor (R-factor), while at high Re numbers, rectangular winglets prevail. Compared to trapezoidal winglet configurations, rectangular winglets enhance the Nu by 6.49%-6.67%. Among the various attack angle configurations, the 45° rectangular winglets exhibit the most favorable overall heat transfer performance, with a Nu increase of 12.07% compared to standard H-type fins. It is observed that installing more than two VGs can result in an R-value less than 1, suggesting that excessive VG placement in H-type fin heat exchangers is not advisable. Finally, the Non-dominated Sorting Genetic Algorithm II (NSGA-II) is employed to identify Pareto optimal points, which exhibit lower resistance and enhanced convective heat transfer. The Pareto optimal set, in comparison to different attack angle CFD data, shows a 31.27% reduction in resistance and a 15.72% increase in Nu.

Soil-dependent stochastic optimization of base isolation system with negative stiffness inerter damper subjected to non-stationary seismic excitation

The negative stiffness inerter damper (NSID) has demonstrated excellent performance in mitigating vibrations in frame structures. However, analysis and optimization of BIS-NSID considering both soil conditions and non-stationary seismic excitation are rarely reported. This study proposes a multi-objective performance optimization procedure (MPOP) to investigate the optimal parameters of BIS-NSID under non-stationary seismic excitation considering firm, medium and soft soil conditions. The non-stationary seismic excitation is expressed as Clough-Penzien model with time modulation function. The analytical solution for the root mean square responses of BIS-NSID are derived considering non-stationary seismic excitation. The pareto optimal fronts is utilized to minimize the BIF displacement and superstructure acceleration. The optimal design is further validated under artificial seismic records. The results illustrate the validity of the MPOP and highlight the advantage of the BIS-NSID for all soil conditions. The demand for inertance and damping values are greater in soft soil conditions, compared to firm and medium soil condition.

A greedy post-processing strategy for multi-objective performance optimization of general single-server finite queueing networks

A greedy post-processing strategy for multi-objective performance optimization of general single-server finite queueing networks

Surrogate multi-objective optimization of missile RCS performance in hypersonic rarefied flow of the near space

This study focuses on optimizing the lateral jet efficiency of THAAD-like (Terminal High Altitude Area Defense) missiles operating under hypersonic rarefied flow conditions. We employ the DSMC-QK algorithm to simulate the three-dimensional lateral jet flow field, accounting for thermochemical non-equilibrium effects. The analysis investigates how the force/momentum amplification coefficient varies with the angle of attack, jet pressure ratio, jet Mach number, and jet gas composition. Subsequently, we develop an artificial neural network (ANN) proxy model using the pyrenn toolbox, achieving an average prediction error of 0.866% and a maximum error of 1.60%. Utilizing this ANN model, we perform single- and multi-objective optimizations with a genetic algorithm to determine the optimal jet parameters. The results reveal that in multi-objective optimization, the proportion of helium in the jet gas composition increases, leading to a slight reduction in the force amplification coefficient but a substantial 61.4% decrease in the mass flow rate. This demonstrates that a judicious selection of jet gas composition can significantly reduce mass flow while maintaining high jet efficiency, thus achieving efficient lateral jet control.

Model construction and multi-objective performance optimization of a biodiesel-diesel dual-fuel engine based on CNN-GRU

To tackle the challenges of low efficiency and excessive NOx emissions in biodiesel-diesel dual-fuel engines, this paper presents an optimization method using the Pareto-multiple objective snake optimizer (Pareto-MOSO) to drive a convolutional neural network-gated recurrent unit (CNN-GRU). In the Pareto-MOSO, brake specific fuel consumption (BSFC), NOx, Soot, and CO are optimized by changing the engine control parameters, including injection timing (Tinject), engine speed (n), biodiesel blending ratio, torque (Ttorque), exhaust gas recirculation (EGR rate), intake pressure (Pin), and rail pressure (Prail). The experiment generates training data for CNN-GRU and verifies the accuracy of the Pareto-MOSO optimization results. The findings suggest that the optimized engine exhibits a balanced correlation between economy and emissions under propulsion characteristics. Furthermore, the NOx emissions are all in accordance with the IMO Tier III emission regulations. Under the rated condition, Scheme 2 demonstrates a significant reduction of 76.57 % in NOx emissions. However, this optimization has resulted in a 1.63 % increase in BSFC compared to its pre-optimized state. Therefore, the adoption of suitable control strategies proves advantageous in addressing the trade-off between economy and emissions of engines.

Multi-objective optimization of methanol reforming reactor performance based on response surface methodology and multi-objective particle swarm optimization coupling algorithm for on-line hydrogen production

The utilization of high-temperature exhaust to drive methanol steam reforming for on-line hydrogen production in engines is an effective approach to solving hydrogen storage and transportation challenges. In this work, the response surface methodology (RSM) is combined with the multi-objective particle swarm optimization (MOPSO) algorithm, and is first utilized for multi-objective optimization of the performance of a methanol reforming reactor. Regression models for optimization objectives are developed based on RSM, and the reliability and significance of regression models are determined through analysis of variance. The influence mechanism of the interaction between design variables on optimization objectives is elucidated by visual analysis of the response surface. A multi-objective optimization of reactor performance is performed utilizing the MOPSO algorithm, and the Pareto optimal solution is obtained from the Pareto optimal frontier based on the technique for order preference by similarity to ideal solution method. Optimization results demonstrate that the Pareto optimal solution is obtained at an exhaust temperature of 585.21 K, an exhaust flow rate of 0.0097 kg·s−1, a steam to methanol ratio of 1.48, and a weight hourly space velocity of 1.2. The hydrogen yield, methanol conversion and carbon monoxide selectivity corresponding to the Pareto optimal solution are 68.53 mol·h−1, 0.83 and 0.011, respectively. Under optimal operating parameters, the methanol reforming reactor achieves 21.92 % exhaust energy recovery.

Parametric design and multi-objective optimization of daylight performance in gallery skylight systems: A case study on the high museum expansion

In the realm of museum design, the meticulous planning of natural lighting plays a vital role in enhancing energy efficiency, improving the indoor atmosphere, and preserving and showcasing artifacts. A significant challenge is to ensure visitors' visual comfort while minimizing excessive natural light to prevent damage to exhibits. This study takes Renzo Piano's High Museum Expansion as a case study and employs the medium sensitivity exhibit standards from IESNA as the evaluation criteria. During the preliminary design phase, computer modeling was utilized, daylighting simulations were carried out using the Ladybug tool, and the Octopus genetic algorithm plugin was applied to investigate, optimize, and evaluate the performance of the point skylight system in museum interiors. The findings indicate that optimized point skylight systems not only block direct sunlight, reducing indoor glare, but also ensure that the Annual Cumulative Illuminance (ACI) on interior surfaces stays below 480,000 lx, while the annual effective Useful Daylight Illuminance (UDI) is increased to 75 %. In terms of design parameters, a skylight height of 0.6 to 0.7 m, combined with a shading panel to horizontal angle of 65 to 70 degrees, is found to be effective in enhancing indoor lighting conditions.

Multi-objective optimization of energy and daylight performance for school envelopes in desert, semi-arid, and mediterranean climates of Iran

Considering global warming as a critical challenge to human life, performance optimization of the building envelope can play a noticeable role to reduce a building's environmental footprint. Meanwhile, shading systems are considered as sustainable passive solutions to contribute to building energy efficiency, and daylight control in early-stage design. Nonetheless, a paucity of research has examined the various shading strategies in diverse climates, particularly arid desert and steppe regions, in conjunction with educational buildings. The present study aims to evaluate the role of fixed exterior shading systems (FESSs), and window-to-wall ratio (WWR) on building’ thermal and daylight performance in Iran with desert, semi-arid, and Mediterranean climates. Pareto Frontier and weighted-sum method for multi-objective optimization, and sensitivity analysis were used to study the optimum solutions, and the relationship between the design variables and the performance metrics. Three objective metrics including Spatial Daylight Autonomy (sDA), Annual Sunlight Exposure (ASE), and Energy Use Intensity (EUI), were defined as performance metrics. Shadings include vertical louver, horizontal louver, light shelf, overhang, and egg-crate. The results showed that EUI was reduced via FESS-integrated façade by 25.22%, 20.84%, 19.14%, 14.06%, and 13.47%, in Yazd, Bushehr, Kerman, Rasht, and Mashhad as selected studied cities, respectively. Based on ASE, the horizontal louver surpasses the other systems by 100% ASE value reduction in all climate zones except for Rasht. sDA level is reduced in all climate zones considering five studied FESSs excluding horizontal louver in Yazd and Rasht and overhang in Mashhad by 100% sDA level. Building energy performance simulation is validated by ASHRAE140-2020.

Multi-objective prediction and optimization of performance of three-layer latent heat storage unit based on intermittent charging and discharging strategies

An intermittent heat charging and discharging strategy is proposed for on-demand thermal utilization in a three-layer latent heat storage unit filled with nanoparticle-enhanced phase change materials. To optimize the utilization ratio of phase change materials, and the stored and released thermal exergy amounts, a multi-objective prediction and optimization methodology combining orthogonal experimental design, range and variance analyses, multi-nonlinear regression models, and non-dominated sorting genetic algorithm-II is introduced while considering the variables of nanoparticle concentration, heat transfer fluid velocity, and intermittent time interval. Results show that the time interval presents the most significant influence. Multi-nonlinear regression models for the above three variables are established with determination factors of 0.9871, 0.9625, and 0.9253, respectively. The ultimate optimal results are 0.8, 57094.03 J, and 43066.73 J, achieved at the three variables of 44.37 min, 0.38 m s−1 and 8.99%, respectively. The maximum verification error of 5.11% indicates the reliability of this methodology. The methodology aims to enhance the overall performance of the three-layer latent heat storage system by mitigating the constraints associated with single-performance optimization.

Multi-objective optimization of residential building energy consumption, daylighting, and thermal comfort based on BO-XGBoost-NSGA-II

The energy consumption, daylighting, and thermal comfort of buildings directly affect the three key goals of residents. However, there is little research on the optimization of energy consumption, daylighting, and thermal comfort in residential buildings in China. Therefore, this study proposes an optimization framework that combines Bayesian optimization with extreme gradient boosting trees (BO-XGBoost) and non-dominated genetic algorithm-II (NSGA-II) to study the multi-objective optimization of residential building performance. This paper first uses Grasshopper to simulate and obtain a dataset through Latin hypercube sampling (LHS). BO-XGBoost is used to establish the regression relationship between building envelope design parameters and residential building performance. Then, the obtained regression model is used as the fitness function of NSGA-II to get the Pareto optimal solution set. Finally, the ideal point method is used to obtain the optimal combination of building envelope structure parameters for residential buildings. Taking a residential building in a hot summer and cold winter area as an example, the effectiveness of this method is verified. The results show that (1) BO-XGBoost has excellent predictive performance, with R2 values of 0.997, 0.960, and 0.994 for energy consumption, thermal comfort, and daylighting, respectively. (2) The proposed BO-XGBoost-NSGA-II can effectively achieve multi-objective optimization. Compared with the initial scheme of the case building, energy consumption is reduced by 44.1%, thermal comfort index is reduced by 10.9%, and daylighting performance is improved by 1.7%. Therefore, the proposed method can effectively optimize the performance goals of residential buildings and provide practical ideas for similar problems.

Multi-objective optimization of heat charging performance of phase change materials in tree-shaped perforated fin heat exchangers

To enhance the convective heat transfer and overall thermal performance of horizontal latent heat storage (LHS) units, an innovative tree-shaped fin structure with perforations was introduced. A numerical simulation study examined the impact of perforation layers on the unit's overall performance. For a comprehensive analysis of the perforations' cumulative effect on the LHS unit, the Response Surface Methodology (RSM) was utilized to assess the influence of variations in perforation diameter and number on material consumption and heat exchange targets, deriving predictive correlations. The results indicated that compared to a non-perforated structure, the maximum improvement was observed with three layers of perforations, reducing material usage by 0.93%, while increasing the Nusselt number (Nu) and heat exchange by 10.19% and 36.51%, respectively. Moreover, the average temperature of the phase change material (PCM) in one and three perforated layers was higher than in the non-perforated tree-shaped fins, with complete melting times reduced by 5.37% and 2.51%, respectively. RSM results showed a negative linear correlation between increased perforation diameter and number with reduced material consumption, with the number of perforations having a more significant impact on heat exchange than their diameter. The Pareto optimal points obtained using the Non-dominated Sorting Genetic Algorithm II (NSGA-II) demonstrated lower material usage and higher heat exchange. Compared to the non-perforated tree-shaped fins, the Pareto-optimized solutions reduced material consumption by 2%–2.68% and increased heat exchange by 79.42%–94.45%. Analysis of the flow field, temperature field, and phase change process revealed that perforations significantly enhanced the mixing of PCM in different areas of the tree-like fins and secondary flow, with temperatures near the perforations increasing by 5–7K and flow velocity by 3–4 times, significantly promoting convective heat transfer.

Performance prediction and multi-objective optimization for the Atkinson cycle engine using eXtreme Gradient Boosting

In this study, substantial efforts have been done to enhance both the economy and nitrogen dioxide (NOx) emission characteristics of an Atkinson cycle engine (ACE). The integrated simulation model was meticulously formulated and calibrated utilizing GT-Power software based on test data, and four machine learning (ML) models were developed to predict the performance of the ACE. On this basic, the improved multi-objective hybrid particle swarm optimization (IMHPSO) algorithm was proposed to conduct comprehensive optimization of ACE performance. The results highlight the superior prediction performance and generalization ability of the eXtreme Gradient Boosting (XGBoost) model, attaining an average coefficient of determination (R2) of 0.9978 and 0.9966 on the train and test sets, respectively. The IMHPSO algorithm exhibited efficient convergence, uniform distribution and rich diversity across five benchmark test problems, effectively mitigating issues related to local optima and premature convergence. The Optimal Brake Specific Fuel Consumption (BSFC) solution is chosen as the preferred optimized scheme and realizes significant reductions of 8.5 % for BSFC and 96.1 % for NOx emissions, which experiences a remarkable increase of over 1300 times in computational efficiency compared to the GT-Power model. These findings provide theoretical basis, data support and novel insights for multi-objective optimization of ACE performance.

Multi-objective performance optimization & thermodynamic analysis of solar powered supercritical CO2 power cycles using machine learning methods & genetic algorithm

The present study is focused on multi-objective performance optimization & thermodynamic analysis from the perspectives of energy and exergy for Recompression, Partial Cooling & Main Compression Intercooling supercritical CO2 (sCO2) Brayton cycles for concentrated solar power (CSP) applications using machine learning algorithms. The novelty of this work lies in the integration of artificial neural networks (ANN) and genetic algorithms (GA) for optimizing the performance of advanced sCO2 power cycles considering climatic variation, which has significant implications for both the scientific community and engineering applications in the renewable energy sector. The methodology employed includes thermodynamic analysis based on energy, exergy & environmental factors including system performance optimization. The system is modelled for net power production of 15 MW thermal output utilizing equations for the energy and exergy balance for each component. Subsequently, thermodynamic model extracted dataset used for prediction & evaluation of Random Forest, XGBoost, KNN, AdaBoost, ANN and LightGBM algorithm. Finally, considering climate conditions, multi-objective optimization is carried out for the CSP integrated sCO2 Power cycle for optimal power output, exergy destruction, thermal and exergetic efficiency. Genetic algorithm and TOPSIS (technique for order of preference by similarity to ideal solution), multi-objective decision-making tool, were used to determine the optimum operating conditions. The major findings of this work reveal significant improvements in the performance of the advanced sCO2 cycle by 1.68 % and 7.87 % compared to conventional recompression and partial cooling cycle, respectively. This research could advance renewable energy technologies, particularly concentrated solar power, by improving power cycle designs to increase system efficiency and economic feasibility. Optimized advanced supercritical CO2 power cycles in concentrated solar power plants might increase renewable energy use and energy generation infrastructure, potentially opening new research avenues.

Conception and thermo-economic performance investigation of a novel solid oxide fuel cell/ gas turbine/Kalina cycle cascade system using ammonia-water as fuel

In order to develop a zero-carbon, high-efficiency and low-cost energy system, the conception of an innovative ammonia-water fueled solid oxide fuel cell (SOFC) - gas turbine (GT) - Kalina cycle (KC) cascade power system is proposed. The performance superiority of novel cascade system is verified by comparing with SOFC system using anode off-gas recycle (AOR). The influence of system key operation parameters including fuel ammonia mass fraction on novel system total energy efficiency and levelized cost of electricity (LCOE) is revealed. The rapid prediction and multi-objective optimization of novel system thermo-economic performance are conducted based on artificial neural network - multi-objective Harris hawks optimizer (ANN-MHHO). The results show that the total energy efficiency and LCOE of novel ammonia-water fed SOFC-GT-KC system are 12.92% higher and 8.91% lower than the SOFC system using AOR under preliminary condition, respectively. Parametric analysis reveals that fuel ammonia mass fraction presents a more significant effect on novel system thermo-economic performance than cell stack fuel utilization ratio. A high cell stack fuel utilization ratio of more than 0.64 is not conducive to improving novel system economic performance. The novel system LCOE reaches the minimum of 0.13807 $/kWh at SOFC operation pressure of 9.5 bar. The multi-objective optimization results demonstrate that the novel system total energy efficiency is increased by 1.14% and LCOE is reduced by 1.23% under optimal condition compared with preliminary condition. The novel system LCOE is more sensitive to ammonia price than interest rate.

Experimental analysis and multi-objective optimization of flame dynamics and combustion performance in methane-fueled slit-type combustors

Achieving a high wall temperature and wide flame stability limit is of extreme significance for the maximization of power generation in hydrocarbon fuel-driven micro-thermophotovoltaic and thermoelectric devices. For this, detailed experimental investigations are conducted on variable channel height combustors, with the emphasis on analyzing the flame dynamics and thermal performance. A preliminary understanding of the multiple flame morphologies in the combustors is developed by varying the input working conditions and wall materials. The basic types of unstable flames are clarified. The pinch-off phenomenon of a weak flame in a slit combustor is observed for the first time, and the cell tends to decrease and emerge with an increase in the flow rate. The combustible range is shown to be extended with increasing the channel height, while it exhibits a non-monotonic changing trend with the combustor thermal conductivity which highlights the importance of the balance between heat transfer due to the flame-wall coupling and heat losses. Furthermore, a Kriging-NSGA-II optimization model is also developed to obtain the optimal characteristics in terms of radiation efficiency, standard deviation and volume power density. A Pareto-optimal front solution is determined to identify three distinct regions of thermal conductivities, which is crucially useful in guiding the design of practical micro-power systems based on different working requirements.

Constructing a Multi-Objective Optimization Model for Engineering Projects Based on NSGA-II Algorithm under the Background of Green Construction

In the context of Sustainability Development (SD), Green Construction (GC) has become a key direction for optimizing engineering project objectives. In order to improve the management ability of project engineering in GC, an improved NSGA - II algorithm was used in this study to establish a multi-optimization model for engineering projects. In this process, the hill climbing is introduced to improve the search ability of NSGA - Ⅱ algorithm. Finally, a Multi-Objective Optimization (MOP) model with strong convergence and distribution was obtained. In subsequent validation experiments, the total construction period of the engineering project MOP model based on the improved NSGA - II algorithm was between 190 and 234days. The total cost ranges from 171,473 to 20,461,800 yuan. Its total mass ranges from 90.41% to 92.19%. Its total safety is between 91.30% and 99.32%. The total environment is between 144.54 and 193.58. Its total resources range from 86.21% to 99.91%. The cost of improving the NSGA-II algorithm is 500300 yuan lower than that of the NSGA-II algorithm, with a resource target increase of 0.4% and an environmental target increase of 4.33%. The iteration curves of the improved NSGA - II algorithm in terms of duration, cost, and environmental objective function are lower than those of the NSGA - II algorithm. Overall, the improved NSGA - II algorithm has better MOP performance, can obtain better Pareto solutions, and has better performance.

Multiobjective optimization of daylighting, energy, and thermal performance for form variables in atrium buildings in China's hot summer and cold winter climate

Multiobjective optimization of daylighting, energy, and thermal performance for form variables in atrium buildings in China's hot summer and cold winter climate