Sequential Genetic Algorithm Research Articles

Optimization of process parameters in laser cladding multi channel forming using MVBM-NSGA-II method

ABSTRACT Laser cladding multi-channel forming is a complex, nonlinear process characterized by multi-parameter coupling. To determine the technology parameters, an optimization method of multivariate black-box prediction model (MVBM) matching non-dominated sequential genetic algorithm (NSGA-II) was constructed. Multi-channel LC orthogonal experiments were designed. The MVBM was established with process technology parameters as inputs and characteristic parameters reflecting the quality (surface hardness, dilution rate) as responses. The set of optimal solutions for technology parameters by the NSGA-II. The experimental results showed that, under the optimum parameters, the average hardness of the optimized increased by 6.1%, and the dilution rate was reduced by 49.8%. The dilution rate was maintained within the optimal range. The study’s findings can provide theoretical support for optimizing LC multi-objective technology parameters and improving coating quality control.

Examination of convective heat transfer and entropy generation performance in twisted elliptical tubes using response surface method

In this paper, the convective heat transfer (CHT) and flow properties of twisted elliptical tubes (TETs) are numerically simulated within the range of Reynolds numbers (Re) 710 to 4790. The effects of design parameters such as twist ratio (P/D), aspect ratio (A/B), and Re on the CHT and flow resistance properties of TETs are highlighted. The response surface method (RSM) is used to research the effect of interaction between design parameters on the performance of CHT in TETs and establish the relationship between the objective function, i.e., Nusselt number (Nu), friction factor (f), and design parameters. The objective function fitted by RSM are optimized by the non-dominated sequential genetic algorithm (NSGA-II) to obtain the Pareto optimal solution. Besides, the entropy generation minimization principle explores entropy generation characteristics to reveal the mechanism of CHT enhancement in TETs. The results show that the CHT performance, flow resistance, and total entropy generation in TET increase, decrease, and increase with the growth of Re, respectively. They all increase with the reduction of P/D or the growth of A/B. The maximum values of Nu and pressure drop (ΔP) occur when A/B = 2.16 and P/D = 11.90; the minimum value of total entropy generation is reached when A/B = 1.34 and P/D = 26.33; and the Pareto optimal solution corresponds to the best combination of design parameters of Re = 3980, P/D = 13.32, and A/B = 1.82.

A new hybrid parallel genetic algorithm for multi-destination path planning problem

This paper proposes a new parallel approach of multi objective genetic algorithm for path planning problem. The main contribution of this work is to reduce the population size that effect in decreasing processing times of finding the optimum path for multi destination problem. This is achieved by combining the local population of island parallel approach and global population of global parallel approach. Various experiments have been conducted to evaluate the new hybrid parallel genetic algorithm (HPGA) in solving multi-objective path planning problems. Three different test areas with 2 destinations were used to assess the performance of HPGA. Furthermore, this work compares HPGA and sequential genetic algorithm (SeqGA), as well as compared to other existing parallel genetic algorithm (GA) methods. From experimental results show that proposed HPGA outperform others, in term of processing time i.e., up to 3.6 times speedup faster, and lowest GA parameter values. This proposed HPGA can be utilized to design robots with fast and consistent path planning, especially with various obstecles.

Suppression of stimulated Brillouin scattering effect in 2 μm narrow linewidth Tm3+-doped fiber amplifier based on phase-modulated waveform design

In this paper, we investigate the optimal design of phase modulation waveforms using the theoretical model of Tm3+-doped fiber amplifiers. The proposed design suppresses the SBS effect in the 2 μm-band Tm3+-doped fiber amplifiers with the help of a reference point-based multi-device non-dominated sequential genetic algorithm (NSGA-III). At the same time, synthesis efficiency is ensured. Different from the previous use of conventional multi-objective genetic algorithms to suppress the SBS effect in narrow linewidth amplifiers in the 1 μm band, the maximum SBS threshold is 121 W at a signal linewidth of 496 MHz with a more uniform distribution of Pareto solutions under this scheme. The signal linewidth of 680.61 MHz increases the SBS threshold by almost 20 W and the operating speed by several times compared to a conventional multi-objective genetic algorithm at different modulation depths, modulation frequencies and modulation periods. The threshold enhancement factor of the SBS signal with optimal modulation is 2.12 times that of the sinusoidal signal with an extended bandwidth of 700 MHz. The 2 μm laser obtained by this scheme enables more accurate and longer distance ranges in coherent detection.

Performance research and optimization of marine dual-fuel engine based on RSM and NSGA-II

ABSTRACT In order to improve the economic efficiency of the ship when sailing and reduce the engine optimization cost. We propose the Response Surface Methodology (RSM) combined with NSGA-II to optimize the engine parameters. First, a simulation model of a marine four-stroke dual-fuel engine is established in AVL-BOOST software. Then, control parameters such as engine speed, exhaust valve opening (EVO) and compression ratio (CR) are planned by design of experiments. The response surface model was established in Design-Expert software. The significant influence of control parameters on performance parameters was studied by analysis of variance (ANOVA). Finally, with the output power, indicated fuel consumption rate and nitrogen oxide emissions as the optimization objectives. Non-dominated Sequential Genetic Algorithm (NSGA-II) is used to optimize the parameters to improve engine performance and reduce emissions. The results show that the established response surface model has good prediction accuracy. The response surface model visualizes the mathematical relationship between the control parameters and the optimization targets. The ANOVA results show that engine speed, EVO and CR have significant effects on engine performance and emissions. The optimization results show that the engine speed is 793 rpm, the EVO is 145°CA, and the CR is 12.3. Compared to standard settings, the optimized data shows a 3.4% increase in power, a 0.3% reduction in ISFC, and a 6.2% reduction in nitrogen oxide (NOx) emissions. The combination of response surface analysis and NSGA-II algorithm to optimize engine performance and emissions is thereby a feasible method.

On the behavior of parallel island models

Parallel island models (PIMs) are used to enhance the performance of evolutionary algorithms (EAs). In PIMs, each island executes an EA for evolving its local population, and periodically individuals migrate to neighborhoods synchronously or asynchronously. Neighborhoods are organized through a topology of communication, and migration policies guide individual exchange. This work explores migration policies over different communication topologies in synchronous and asynchronous PIMs to improve the speed-up and accuracy of genetic algorithms (GAs). The aim is to explain such models’ adequacy from a general perspective, attempting to answer questions such as the best way to implement GAs in PIMs. To reach this goal, the quality of the solutions and the running time provided by proposed PIMs are evaluated over four NP-hard problems of different combinatorial nature: reversal and translocation evolutionary distance, task mapping and scheduling, and N-Queens. The experiments show that tuning the parameters of the breeding cycle and migration policies is vital to guarantee that all proposed PIMs reach good speed-ups and more accurate solutions than the sequential GA. In addition, experiments ratify that synchronous models provide the best solutions, while asynchronous models deliver the best speed-ups. Finally, the results show that no model provides either better speed-up or accuracy in general since, as for sequential models, the nature of the problem define which would be the best-adapted PIM.

Effect of foundation embedment ratio in suppressing seismic-induced vibrations using optimum tuned mass damper

Tuned mass damper (TMD), which is used in structures to mitigate the dynamic responses caused by environmental loads, is tuned according to the dominant frequency of the structures. The dominant frequency of structure built on soil with low stiffness is significantly affected by the soil-structure interaction (SSI). In previous studies, only the surface foundation case is considered to investigate the design parameters of TMD considering SSI effects and the embedment depth of the foundation is neglected in these studies. Thus, this paper investigates the optimization of TMD for mitigation of responses of a high-rise building considering various depths of embedment and soil properties. The optimum parameters of TMD are investigated using five different optimization algorithms (i.e., simulated annealing, sequential quadratic programming, genetic algorithm, pattern search, and particle swarm optimization). The objective function minimizes the peak acceleration amplitude at the top of the structure. The effectiveness of optimum TMD is verified to use 20 near-fault pulse-like ground motions. The comparisons show that the embedment ratio of the foundation and the soil properties significantly affect the optimization of TMD.

Investigation of blueberry juice fermentation by mixed probiotic strains: Regression modeling, machine learning optimization and comparison with fermentation by single strain in the phenolic and volatile profiles

Three strains, including L. fermentum, L. plantarum and S. thermophilus, were combined to ferment blueberry juice. Through the sequential simplex lattice mixture design, regression modeling and genetic algorithm optimization, it was found that the combination of S. thermophilus with either L. fermentum or L. plantarum weakened the capacity of Lactobacillus strains to enrich phenolics, and the combinations of these strains had no synergistic effect of synthesizing lactic acid. The resulting optimal inoculation proportion to enrich phenolics was the mixed L. fermentum and L. plantarum at 0.5:0.5. After fermentation for 48 h, total phenolic, ferulic acid, rutin, and quercetin-3-rhamnoside of mixed fermented samples were 82.19 %, 15.22 %, 79.08 % and 98.59 % higher than the unfermented juice, and their contents were all highest among the fermented samples. Moreover, the samples fermented by mixed strains possessed higher amounts of 3-methyl-1-butanol, 2-methyl-1-propanol, 2-heptanone and 2-pentanone than samples fermented by L. fermentum, S. thermophilus and unfermented samples.

Multi-objective macrogeometry optimization of gears: Comparison between sequential quadratic programing and genetic algorithm

Deciding a gear pair’s macrogeometry parameters requires consideration of the factors such as cost of production, gear strength, and noise contributing parameters. A practical engineering problem is encountered when best possible gear geometry is needed for a fixed center distance to achieve the conflicting objectives. Further, constraints of practical utility on input parameters need to be taken into consideration for deciding the design space. This work represents comprehensive compilation of formulations of these often-conflicting objectives and constraints as non-linear functions and selection of an appropriate optimization technique to achieve a feasible design through a case study of a helical gear pair in high load carrying capacity mesh. For comparison of the quality of results and computational efficiency between gradient based and gradient free algorithms, Sequential Quadratic Programming and Genetic algorithm were used. The differences in the formulations of the objective functions for two algorithms were analyzed and addressed. Quadratic mutation rates and random crossovers were used to prevent premature convergence of Genetic Algorithm. Significant possible improvement in the base design was observed with both the algorithms.

Optimization-based biomechanical lifting models for manual material handling: A comprehensive review.

Lifting is a main task for manual material handling (MMH), and it is also associated with lower back pain. There are many studies in the literature on predicting lifting strategies, optimizing lifting motions, and reducing lower back injury risks. This survey focuses on optimization-based biomechanical lifting models for MMH. The models can be classified as two-dimensional and three-dimensional models, as well as skeletal and musculoskeletal models. The optimization formulations for lifting simulations with various cost functions and constraints are reviewed. The corresponding equations of motion and sensitivity analysis are briefly summarized. Different optimization algorithms are utilized to solve the lifting optimization problem, such as sequential quadratic programming, genetic algorithm, and particle swarm optimization. Finally, the applications of the optimization-based lifting models to digital human modeling which refers to modeling and simulation of humans in a virtual environment, back injury prevention, and ergonomic safety design are summarized.

A Two-Stage Multi-Objective Genetic Algorithm for a Flexible Job Shop Scheduling Problem with Lot Streaming

The work in this paper is motivated by a recently published article in which the authors developed an efficient two-stage genetic algorithm for a comprehensive model of a flexible job-shop scheduling problem (FJSP). In this paper, we extend the application of the algorithm to solve a lot streaming problem in FJSP while at the same time expanding the model to incorporate multiple objectives. The objective function terms included in our current work are the minimization of the (1) makespan, (2) maximum sublot flowtime, (3) total sublot flow time, (4) maximum job flowtime, (5) total job flow time, (6) maximum sublot finish-time separation, (7) total sublot finish-time separation, (8) maximum machine load, (9) total machine load, and (10) maximum machine load difference. Numerical examples are presented to illustrate the greater need for multi-objective optimization in larger problems, the interaction of the various objective function terms, and their relevance in providing better solution quality. The ability of the two-stage genetic algorithm to jointly optimize all the objective function terms is also investigated. The results show that the algorithm can generate initial solutions that are highly improved in all of the objective function terms. It also outperforms the regular genetic algorithm in convergence speed and final solution quality in solving the multi-objective FJSP lot streaming. We also demonstrate that high-performance parallel computation can further improve the performance of the two-stage genetic algorithm. Nevertheless, the sequential two-stage genetic algorithm with a single CPU outperforms the parallel regular genetic algorithm that uses many CPUs, asserting the superiority of the two-stage genetic algorithm in solving the proposed multi-objective FJSP lot streaming.

Multiobjective Macrogeometry Optimization of Gears: Comparison between Sequential Quadratic Programing and Genetic Algorithm

Multiobjective Macrogeometry Optimization of Gears: Comparison between Sequential Quadratic Programing and Genetic Algorithm

Adequacy Evaluation of Composite Power Systems Using an Evolutionary Swarm Algorithm

The generation and transmission capacities of many power systems in the world are significantly increasing due to the escalating global electricity demand. Therefore, the adequacy evaluation of power systems has become a computationally challenging and time-consuming task. Recently, population-based intelligent search methods such as Genetic Algorithms (GAs) and Binary Particle Swarm Optimization (BPSO) have been successfully employed for evaluating the adequacy of power generation systems. In this work, the authors propose a novel Evolutionary Swarm Algorithm (ESA) for the adequacy evaluation of composite generation and transmission systems. The random search guiding mechanism of the ESA is based on the underlying philosophies of GAs and BPSO. The main objective of the ESA is to find out the most probable system failure states that significantly affect the adequacy of composite systems. The identified system failure states can be directly used to estimate the system adequacy indices. The proposed ESA-based framework is used to evaluate the adequacy of the IEEE Reliability Test System (RTS). The estimated annualized and annual adequacy indices such as Probability of Load Curtailments (PLC), Expected Duration of Load Curtailments (EDLC), Expected Energy Not Supplied (EENS) and Expected Frequency of Load Curtailments (EFLC) are compared with those obtained using Sequential Monte Carlo Simulation (SMCS), GA and BPSO. The results show that the accuracy, computational efficiency, convergence characteristics, and precision of the ESA outperform those of GA and BPSO. Moreover, compared to SMCS, the ESA can estimate the adequacy indices in a more time-efficient manner.

Towards a model-independent reconstruction approach for late-time Hubble data

Gaussian processes offers a convenient way to perform nonparametric reconstructions of observational data assuming only a kernel which describes the covariance between neighbouring points in a data set. We approach the ambiguity in the choice of kernel in Gaussian processes with two methods — (a) approximate Bayesian computation with sequential Monte Carlo sampling and (b) genetic algorithm — and use the overall resulting method to reconstruct the cosmic chronometers and supernovae type Ia data sets.The results have shown that the Matérn( ν = 5/2 ) kernel emerges on top of the two-hyperparameter family of kernels for both cosmological data sets. On the other hand, we use the genetic algorithm in order to select a most naturally-fit kernel among a competitive pool made up of a ten-hyperparameters class of kernels. Imposing a Bayesian information criterion-inspired measure of the fitness, the results have shown that a hybrid of the Radial Basis Function and the Matérn( ν = 5/2 ) kernel best represented both data sets. The kernel selection problem is not totally closed and may benefit from further analysis using other strategies to resolve an optimal kernel for a particular data set.

Sequential Multi-objective Genetic Algorithm

Many of the real-world issues have multiple conflicting objectives that the optimization between contradictory objectives is very difficult. In recent years, the Multi-objective Evolutionary Algorithms (MOEAs) have shown great performance to optimize such problems. So, the development of MOEAs will always lead to the advancement of science. The Non-dominated Sorting Genetic Algorithm II (NSGAII) is considered as one of the most used evolutionary algorithms, and many MOEAs have emerged to resolve NSGAII problems, such as the Sequential Multi-Objective Algorithm (SEQ-MOGA). SEQ-MOGA presents a new survival selection that arranges individuals systematically, and the chromosomes can cover the entire Pareto Front region. In this study, the Archive Sequential Multi-Objective Algorithm (ASMOGA) is proposed to develop and improve SEQ-MOGA. ASMOGA uses the archive technique to save the history of the search procedure, so that the maintenance of the diversity in the decision space is satisfied adequately. To demonstrate the performance of ASMOGA, it is used and compared with several state-of-the-art MOEAs for optimizing benchmark functions and designing the I-Beam problem. The optimization results are evaluated by Performance Metrics such as hypervolume, Generational Distance, Spacing, and the t-test (a statistical test); based on the results, the superiority of the proposed algorithm is identified clearly.

An advanced LAN model based on optimized feature algorithm: Towards hypertension interpretability

BackgroundThe existing cuffless BP measuring methods still show the inadequacy of the model that utilizes long term dependencies in BP dynamics and come up with the outcome of recurrent calibration. ObjectiveThis study aims to develop a learning-based predictive model for continuous BP measurement with long-term evaluation of the cardiovascular parameter. MethodsTowards hypertension interpretability, the proposed model assists the current as well as the previous status of cardiovascular events based on long short-term memory networks (LSTM) and flexible autoregressive integrated moving average (ARIMA). It is often complicated and challenging to continuously estimate BP from random cardiac events. This paper proposes a new SGFA (Sequential Genetic Feature Algorithm) to tackle the feature optimization problem. ResultsThe experiment was performed on a MIMIC database containing 2780 datafiles of PPG-BP. The proposed model provides the best performance with root mean square error (RMSE) and mean absolute error (MAE) of 1.17 and 1.04 for SBP, whereas 1.06 and 1.02 for DBP. The proposed method has also been evaluated on hypertensive and hypotensive patients. In hypertension condition, estimated SBP and DBP present a good correlation with the true measurements. MAE and RMSE of the estimated SBP are 0.96 and 1.21, whereas, for DBP, it shows 0.42 and 0.57 respectively. Extensive experimentation results confirm that the proposed method delivers a remarkable performance of BP prediction. ConclusionThe proposed work shows a remarkable BP estimation performance compared to the previous inventive methods and signifies insight of various cardiovascular events responsible for BP variation and hypertension interpretability.

Genetic optimization method of ship strong frame based on sequential surrogate model

Objectives The optimization design of a ship strong frame structure under the requirements of common structural rules (CSR) is a complex and time-consuming problem, moreover, it has tremendous constraints so that the feasibility of a design scheme is hard to identify. When adopting the static surrogate assisted evolutionary algorithm to solve this problem, since the approach aims at global accuracy, the prediction of key area will be distorted in the case of small size samples. However, the active-learning surrogate-based approach can solve this bottleneck since it can improve the accuracy of key area sequentially by adding samples to refine the surrogate model purposely. Therefore, this method can obtain better optimization scheme and save computational burden compared with the former one. Methods Aiming at the problem, an optimization method for strong frame based on sequential surrogate assisted genetic algorithm is proposed. Firstly, the constraints on strong frame structure based on CSR are analyzed, and all 675 constraints are reduced to 2 positive constraints according to the type of constraints. Then surrogates for objective function and constraints functions are constructed. Next, feasibility principle based genetic algorithm is adopted to solve it. After optimized solution is obtained, its true response is calculated and surrogates will be updated. In addition, expected feasibility function criterion is applied to update constraints surrogates, the purpose of which is to refine the prediction accuracy at the constraints boundary. Iterate the above procedures for several times, the optimized global feasible solution will be obtained finally. Results It is found that the proposed method is able to get better solution than static surrogate based algorithm with less computation resource and the weight of design area is reduced by 15.55% finally. Conclusions Proposed sequential surrogate based algorithm is superior to static surrogate based algorithm and has good application value in the optimization of ship structure under complex constraints.

Intelligent system for recruitment decision making using an alternative parallel-sequential genetic algorithm

<p>The human resources (HR) manager needs effective tools to be able to move away from traditional recruitment processes to make the good decision to select the good candidates for the good posts. To do this, we deliver an intelligent recruitment decision-making method for HR, incorporating a recruitment model based on the multipack model known as the NP-hard model. The system, which is a decision support tool, often integrates a genetic approach that operates alternately in parallel and sequentially. This approach will provide the best recruiting solution to allow HR managers to make the right decision to ensure the best possible compatibility with the desired objectives. Operationally, this system can also predict the altered choice of parallel genetic algorithm (PGA) or sequential genetic algorithm (SeqGA) depending on the size of the instance and constraints of the recruiting posts to produce the quality solution in a reduced CPU time for recruiting decision-making. The results obtained in various tests confirm the performance of this intelligent system which can be used as a decision support tool for intelligently optimized recruitment.</p>

Flutter Optimization of a Wing–Engine System with Passive and Active Control Approaches

In the present study, the flutter performance of a composite thin-walled beam wing–engine system is optimized by implementing two different control approaches: 1) passive open-loop and 2) active closed-loop control. Sequential quadratic programming and genetic algorithm methods are applied in the optimization process. In the passive control method, variable stiffness is acquired by constructing laminates of thin-walled beam with curvilinear fibers having prescribed paths. The goal is to exploit the desirable fiber paths with improved flutter performance to determine an optimized wing–engine aeroelastic configuration. In the active control strategy, piezo-composite actuators and the linear quadratic Gaussian algorithm are used to improve the flutter characteristics. A novel optimization strategy based on the total energy of the aeroelastic system is introduced and applied in both passive and active control strategies. The minimum total aeroelastic energy is an indication of ideal optimization variables, which leads to optimum flutter performance. The governing equations are formulated based on Librescu’s thin-walled beam theory and Hamilton’s principle. An unsteady aerodynamic model based on incompressible indicial aerodynamics is applied. The governing equations of motion are solved using a Ritz-based solution methodology. Numerical results demonstrate a 16 and 46% improvement in the flutter speed of the wing–engine system using the proposed passive and active control approaches, respectively. The presented results provide valuable information concerning the design of advanced lightweight and high-aspect-ratio aircraft wings with mounted engines in terms of favorable aeroelastic performance characteristics.

Broadband optimization of volumetric sound metadiffusers

Broadband volumetric sound metadiffusers are designed by using gradient-based optimization and by means of multiple scattering theory (MST) employed to evaluate the scattered pressure field by a uniform planar configuration of cylindrical scatterers. The method is illustrated by giving numerical examples for planar configurations of rigid cylinders and thin elastic cylindrical shells situated in the air medium. By rearranging the position of each cylinder, a metadiffuser is optimized to have a high diffusion coefficient. Multiple scenarios are evaluated considering a single frequency and broadband optimization. Single frequency optimization is performed to optimize a configuration of cylinders at a specific wavenumber. Broadband optimization is used to optimize a cylinder configuration for a range of wavenumbers by optimizing the root mean square of diffusion coefficient evaluated at discreet values of normalized wavenumber. MATLAB MultiStart solver combined with parallel computing along with global optimization toolboxes is used to enhance the modeling for optimizing for both scenarios due to its ability to evaluate multiple starting points in parallel, hence being able to find the global optimum. A wide array of metadiffusers is optimized using MATLAB MultiStart and fmincon solvers with the sequential quadratic programming and genetic algorithm for a comparison of performance of these algorithms.