Genetic Algorithm Search Research Articles

Real-time, adaptive microphone subset selection for a 128-elementarray

Some advantages of high-sensor count microphone arrays include gain over random noise, increased multiple-source resolution, and potential increases in effective frequency bandwidth. But in real-time applications, computational capabilities of embedded computing hardware might limit the number data channels that can be processed without data loss. For example, given a specific processing pipeline and sampling rate, perhaps only 24 of 128 channels can be supported in real-time. Therefore, the question of which 24-element subset of the 128 possible elements should be being used to detect and track multiple moving acoustic sources. In this presentation, principles of genetic search algorithms are used to develop a method to adaptively select optimal subsets of sensors as data is being processed. Discussion concerning optimality criteria, types of genetic algorithms, and algorithm performance will be presented as well as results corresponding to a 3-D 128-element array of digital MEMS microphones when multiple types of moving acoustic sources were nearby.

Optimal soft open point placement and open switch position selection simultaneously for power loss reduction on the electric distribution network

Network reconfiguration through the selection of open switch positions on the electric distribution network (EDN) is an effective solution to reduce power loss on the EDN. However, when the switches are opened, there will be no power flow through the corresponding branches, which may not keep the EDN reaching its best operating condition. This paper considers the optimal problem of Soft-Open-Point (SOP) placement and open switch position selection simultaneously on the EDN to reduce the power loss of the system and consider improving the voltage, current profiles, and maintain the radial-topology structure of the EDN. The effect of the number of SOPs on reducing power loss is also considered by increasing the number of SOPs from zero to maximum number of SOPs that still to ensure the radial-topology EDN structure. Growth Optimizer (GO) is first adjusted for finding the optimal solution for the problem of SOP placement and open switch position selection simultaneously. The considered problem and proposed GO method are applied to two test EDNs consisting 33-node and 69-node systems. The GO performance is compared with two well-known genetic algorithm (GA) and cuckoo search algorithm (CSA) and the recently developed northern goshawk optimization algorithm (NGO). The results show that the optimal SOP placement and open switch position simultaneously helps to reduce significantly power loss. The loss reduction compared to the original structure of the two systems in cases of increasing number of SOPs from 0 to 5 is respectively 30.94%, 45.52%, 48.34%, 50.7%, 55.43% and 58.65% for the 33-node system, and 56.16%, 58.43%, 61.99%, 63.78%, 72.80% and 75.05% for 69-node system. The optimal SOP placement and open switch position also helps to improve the voltage quality and capacity of the EDN. In addition, GO reaches the better performance than GA, CSA and NGO for the optimal problem of SOP placement and open switch position selection simultaneously.

A hybrid genetic tabu search algorithm for distributed flexible job shop scheduling problems

The distributed flexible job-shop scheduling problem (DFJSP) is an extension of the flexible job shop scheduling problem, which is a famous NP-complete combinatorial optimization problem. DFJSP consists of three sub-problems, assigning jobs to appropriate factories, scheduling operations to suitable machines, and determining the operation sequence on machines. Due to the complexity of DFJSP, there is only a little research on the problem. There are several benchmark sets proposed to test the performance of methods for solving DFJSP. However, many instances remain unresolved after a long time. This paper proposes a hybrid genetic tabu search algorithm (HGTSA) for DFJSP. The proposed HGTSA combines the global search ability of the genetic algorithm (GA) and the local search ability of the tabu search (TS) well. Two genetic operators are designed based on the critical factory to discrete the population. A new neighborhood structure is introduced into TS to search for more space of neighborhood solutions. To evaluate the performance of HGTSA, this paper compares it with four state-of-the-art algorithms on 69 benchmark instances. The experimental results demonstrate that HGTSA outperforms these comparison algorithms in terms of solution quality and computation efficiency. In particular, HGTSA has found 13 newly upper bounds of these benchmark instances.

Multiple layer kernel extreme learning machine modeling and eugenics genetic sparrow search algorithm for the state of health estimation of lithium-ion batteries

High precision state of health (SOH) estimation of lithium-ion batteries (LIBs) is a research hotspot in battery management system (BMS). To achieve this goal, an improved integrated algorithm based on multiple layer kernel extreme learning machine (ML-KELM) and eugenics genetic sparrow search (EGSS) algorithm is proposed to estimate the SOH of LIBs. First, a kernel version of ML-ELM model is constructed for initial SOH estimation of LIBs. The kernel function parameters are used to simulate sparrow foraging and anti-predatory behaviors, and the parameter optimization process is completed in the proposed EGSS algorithm by iteratively updating the position of sparrows to improve SOH prediction accuracy and model stability. The cycle data of different specifications of LIB units are processed to construct the high-dimensional health feature (HF) dataset and the low-dimensional fusion feature (FF) dataset, and each version of ML-ELM network is trained and tested separately. The numerical analysis of the prediction results shows that the best root mean square error (RMSE) of the comprehensive algorithm for SOH estimation is limited within 0.29%. The results of the multi-indicator comparison show that the proposed algorithm can track the true value stably and accurately with satisfactory high accuracy and strong robustness.

Retracted: Population Economic Transfer Function Model Based on Genetic Tabu Search Algorithm

Retracted: Population Economic Transfer Function Model Based on Genetic Tabu Search Algorithm

Genetic descriptor search algorithm for predicting hydrogen adsorption free energy of 2D material

Transition metal dichalcogenides (TMDs) have emerged as a promising alternative to noble metals in the field of electrocatalysts for the hydrogen evolution reaction. However, previous attempts using machine learning to predict TMD properties, such as catalytic activity, have been shown to have limitations in their dependence on large amounts of training data and massive computations. Herein, we propose a genetic descriptor search that efficiently identifies a set of descriptors through a genetic algorithm, without requiring intensive calculations. We conducted both quantitative and qualitative experiments on a total of 70 TMDs to predict hydrogen adsorption free energy (Delta G_H) with the generated descriptors. The results demonstrate that the proposed method significantly outperformed the feature extraction methods that are currently widely used in machine learning applications.

Research on power system fault prediction based on GA-CNN-BiGRU

Introduction: This paper proposes a power system fault prediction method that utilizes a GA-CNN-BiGRU model. The model combines a genetic algorithm (GA), a convolutional neural network (CNN), and a bi-directional gated recurrent unit network Bidirectional Gated Recurrent Unit to accurately predict and analyze power system faults.Methods: The proposed model employs a genetic algorithm for structural search and parameter tuning, optimizing the model structure. The CNN is used for feature extraction, while the bi-directional gated recurrent unit network is used for sequence modeling. This approach captures the correlations and dependencies in time series data and effectively improves the prediction accuracy and generalization ability of the model.Results and Discussion: Experimental validation shows that the proposed method outperforms traditional and other deep learning-based methods on multiple data sets in terms of prediction accuracy and generalization ability. The method can effectively predict and analyze power system faults, providing crucial support and aid for the operation and management of power systems.

Keyboard Layout Optimization and Adaptation

Since the keyboard is the most common method for text input on computers today, the design of the keyboard layout is very significant. Despite the fact that the QWERTY keyboard layout was designed more than 100 years ago, it is still the predominant layout in use today. There have been several attempts to design better layouts, both manually and automatically. In this paper we improve on previous works on automatic keyboard layout optimization, by using a deep neural network to assist in a genetic search algorithm, which enables the use of a sophisticated keyboard evaluation function that would otherwise take a prohibitive amount of time. We also show that a better choice of crossover routine greatly improves the genetic search. Finally, in order to test how users with different levels of experience adapt to new keyboard layouts, we conduct some layout adaptation experiments with 300 participants to examine how users adapt to new keyboard layouts.

Axial-piston pump optimal multivariable PID controller tunning

The aim of the present work is the synthesis and optimal tunning of a PID controller designed to control the displacement volume of an axial-piston pump intended for hydraulic systems with open circulation. The pump is controlled in real-time through a system developed by the authors based on an electro-hydraulic proportional valve (replacing the conventional hydro-mechanical controller) and an industrial controller. The experimental system was initially used to obtain a data set that is used in a system identification procedure to obtain a model. The obtained state space model was used for the synthesis of the PID controller. The optimal tunning of controller parameters is performed using three modern methods: genetic algorithm, simulated annealing and pattern search. The resulting three PID controllers with different settings depending on the used methods are compared in simulation and experimental conditions. Based on the results, an analysis of the control performance was carried out.

Improved Task Scheduling for Cloud Computing Using Combination of Genetic Algorithm and Tabu Search Algorithm

Abstract: Cloud computing has revolutionized the way computational resources are provisioned and utilized. However, effective task scheduling in cloud environments remains a critical challenge due to the complexity of managing resources and meeting quality of service (QoS) requirements. In this paper, we address the multi-objective task scheduling problem by proposing a novel scheme that combines a hybrid algorithm, HGTSA, to achieve improved convergence rates and overall efficiency. Our main contributions include the development of a fitness function that considers makespan, system capacity, and resource utilization, as well as the implementation of the HGTSA algorithm. Through extensive simulations, we demonstrate the superiority of our proposed scheme, outperforming conventional genetic algorithms and achieving better makespan times. Furthermore, we highlight the future scope of research in leveraging machine learning techniques to further enhance task scheduling processes and optimize decision-making based on workload predictions and historical data.

Application of GA-BP in Displacement Force Inverse Analysis and Mechanical Parameter Inversion of Deep Foundation Pits

Aiming at the defects of various existing displacement inverse analysis methods, using the nonlinear mapping ability of neural network and the global random search ability of genetic algorithm, this paper proposes a displacement inverse analysis method based on optimized Genetic Algorithm- Back Propagation (GA-BP) for deep foundation pit support. The method changes the method that BP algorithm relies on the guidance of gradient information to adjust the network weights, but uses the characteristics of global search of genetic algorithm to find the most suitable network connection rights and network structure, etc. to achieve the purpose of optimization. Firstly, the deformation mechanism of deep foundation pit is analyzed, its failure mode is summarized, and the calculation method of lateral rock and soil pressure is sorted out according to the code. The theory and characteristics of BP neural network and genetic algorithm are discussed, and the method of using genetic algorithm to optimize BP neural network is proposed to improve the prediction accuracy. In view of the shortcomings of GA-BP neural network prediction model in training sample pretreatment and hidden layer structure design, the optimal normalization interval was determined by correlation coefficient regression analysis, and the analytical expression of the number of neurons in hidden layer was derived by statistical principle, and the value range of the optimal number of neurons in single hidden layer was proposed. Combined with the actual engineering, the mechanical parameters inversion and displacement force inverse analysis are performed using this method, and the results show that the optimized GA-BP has higher prediction accuracy compared with BP neural network and GA-BP, and the deviation of the displacement prediction value at each depth is kept within 0.2 mm, the absolute error interval width is 0.07 mm, and the maximum relative error is 1.35% at 4.0 m depth.

Composite disc optimization using hunger games search optimization algorithm

Abstract It has been interesting in industrial applications due to the advantages provided by laminated composite applications. Composite materials are of great importance in design optimization due to the advantages they provide, such as lightweight, high strength value, and design flexibility. Finite element programs and artificial intelligence optimization techniques developed in recent years are effectively used in the design of composite materials. In order to optimize the angular orientations of the composite disc, analyses are made using the finite element program. Particle swarm optimization algorithm (PSO), genetic algorithm (GA), and hunger games search optimization algorithm(HGS) are used to determine the best stacking angle value on the disc plate using the analysis data obtained after the completion of the analyses. When the results obtained were examined, it was determined that the results obtained from particle swarm optimization, genetic algorithm, and hunger games search optimization techniques showed that hunger games search optimization algorithm gives better results as other and HGS could be used for the stacking of composite materials.

Deciding on when to change – a benchmark of metaheuristic algorithms for timing engineering changes

Changes to components, known as engineering changes (ECs), rarely occur on their own. In fact, in complex assembly systems, most ECs are introduced in batches to ensure that changed components match. As a result, to implement ECs optimally, multiple component’s stock must be considered until the change is executed on the EC effectivity date. This problem is known as the EC effectivity date optimisation problem, a variation of the general inventory control problem with deterministic and dynamic demand. As optimisation and monitoring of this problem is computationally expensive, research has suggested to investigate whether metaheuristics can provide adequate support. To fill this research gap, we present the results of a benchmark on basic metaheuristics for EC effectivity date optimisation. To do so, we have compared five common metaheuristics in their basic form (Ant Colony Optimisation, Genetic Algorithm, Particle Swarm Optimisation, Tabu Search, and Simulated Annealing) on a real-world test set. Of the tested algorithms the Genetic Algorithm identified most best solutions and returned good average results for the test cases. However, as its reliability was comparatively low, our research suggests a sequential application of the Genetic Algorithm and Tabu Search.

Single-machine scheduling with periodic maintenance and learning effect

This paper discusses a single-machine scheduling problem with periodic maintenance activities and position-based learning effect to minimize the makespan. To obtain exact solutions of small-scale problems, one new two-stage binary integer programming model is formulated. In addition, a branch and bound algorithm combining boundary method and pruning rules is also proposed. According to the property of the optimal solution, a special search neighborhood is constructed. A hybrid genetic-tabu search algorithm based on genetic mechanism with tabu technique as an operator is proposed to solve medium-scale and large-scale problems. Moreover, to improve the efficiency of genetic algorithm and hybrid genetic-tabu search algorithm, Taguchi method is used for parameter tuning. Furthermore, computational experiments are carried out to compare the efficiency and performance of these algorithms.

Smart charging strategy for electric vehicles based on marginal carbon emission factors and time-of-use price

As the penetration of electric vehicles (EVs) has grown rapidly in recent years, the stability and safety of electric grids are inevitably affected. Uncontrolled EV charging can reduce electric grid efficiency and increase greenhouse gas emissions (GHG). This occurs when EV charging aligns with peak demand, requiring the activation of fossil fuel-powered plants to meet the electricity demand. The time-of-use (TOU) price is a traditional management strategy for smoothing the demand power curve, but it does not consider carbon emissions. This study proposes a smart charging strategy based on an improved local search genetic algorithm (LS-GA) that considers both the TOU price and marginal emission factors (MEF). Over 47,000 charging events from 45 charging stations in Tianjin, a major city in North China, were used to simulate the charging pattern and demand. Four different charging strategies are considered in the study. The results indicate that the proposed smart charging strategy can reduce the cost and carbon emissions by up to 27% and 16% compared with uncontrolled charging, respectively. This study demonstrates the significant potential of using optimization methods in the EV smart charging problem.

A cell marker‐based clustering strategy (cmCluster) for precise cell type identification of scRNA‐seq data

BackgroundThe precise and efficient analysis of single‐cell transcriptome data provides powerful support for studying the diversity of cell functions at the single‐cell level. The most important and challenging steps are cell clustering and recognition of cell populations. While the precision of clustering and annotation are considered separately in most current studies, it is worth attempting to develop an extensive and flexible strategy to balance clustering accuracy and biological explanation comprehensively.MethodsThe cell marker‐based clustering strategy (cmCluster), which is a modified Louvain clustering method, aims to search the optimal clusters through genetic algorithm (GA) and grid search based on the cell type annotation results.ResultsBy applying cmCluster on a set of single‐cell transcriptome data, the results showed that it was beneficial for the recognition of cell populations and explanation of biological function even on the occasion of incomplete cell type information or multiple data resources. In addition, cmCluster also produced clear boundaries and appropriate subtypes with potential marker genes. The relevant code is available in GitHub website (huangyuwei301/cmCluster).ConclusionsWe speculate that cmCluster provides researchers effective screening strategies to improve the accuracy of subsequent biological analysis, reduce artificial bias, and facilitate the comparison and analysis of multiple studies.

The development of computer-based algorithms based on gobang game

Gobang is a worldwide two-player strategy board game, which is popular especially in the Asian region. Recently, the science and technology has been developing fast and artificial intelligence was applied in numerous fields like the board game. Gobang, a strategy game with moderate difficulty, is a suitable example for people to test algorithms and solve board game problems. As a result, many related algorithms were appearing with different advantages and disadvantages. In this work, these already existed algorithms, game tree, minimax search, alpha-beta pruning, genetic algorithm and monte carlo tree search, were discussed and compared. The results and comparison showed that game tree and minimax search had a large number of nodes to calculate to reach a suitable search depth, about 1.00E+12 and 1.29E+14 respectively, which meant they need a long calculating time, while the alpha-beta pruning need to calculate about 2.2E+07 nodes and genetic algorithm only need to calculate about 1.00E+04 nodes, which cost 0.6 seconds for every move. Plus, the monte carlo tree search could reach nearly 100% win rate through self-play, which making gobang algorithm become more refined. Additionally, these algorithms had already made gobang AI powerful with fast move and high win rate, so they also had been applied in many different fields to develop and spread gobang.

A new hybrid algorithm integrating genetic algorithm with Tabu search to solve imbalanced k‐coverage problem in directional sensor networks

AbstractThe target coverage problem is considered as one of the major issues in directional sensor networks (DSNs), which is caused by the nature of these networks, including their limited angle of view. Due to the fault tolerance characteristic of some coverage applications, the target coverage is required to be performed using multiple sensors. This challenge is discussed in the literature under the title of k‐coverage problem. Under certain conditions, the number of sensors may suffer some changes due to various factors such as power depletion of the sensors, sensors' malfunctioning, and harshness of the environment. This can result in unavailability of adequate sensors for providing k‐coverage for all targets. The network suffering from such problem is referred to as under‐provisioned network. This paper was aimed at studying such networks by adopting the network conditions to the real environments. To solve this problem, the present paper proposes a hybrid model integrating the genetic algorithm (GA) and Tabu search (TS). The proposed algorithm generally aimed to identify a subset of sensors with appropriate working directions in order to provide a balanced coverage for all the targets available in the network. In order to evaluate the performance of the algorithm several experiments were conducted and the results have been compared with greedy and learning automat‐abased algorithms. . The results of the experiments show the superiority of the algorithm.

Modelling and optimisation of energy consumption and profit-oriented multi-parallel partial disassembly line balancing problem

Based on the actual requirements of recycling enterprises, this study proposes a multi-parallel partial disassembly line balancing problem (MPPDLBP). Four objectives, the number of shared workstations, workstation load balancing index, energy consumption, and profit, need to be optimised in MPPDLBP. To address the MPPDLBP, this study further constructs a mixed-integer nonlinear programming (MINLP) model and designs a suitable mechanism of encoding and decoding. Meanwhile, partial disassembly is adopted in this study because recycling enterprises aim for low energy consumption and high profits. In addition, the positions of the best values for each objective are defined as the best disassembly levels. Furthermore, this study proposes a genetic and tabu search algorithm (GATS) for optimising the MPPDLBP effectively. The superior performance of the proposed GATS is verified by comparing it with other effective algorithms in existing literature. Finally, this study optimises a hybrid instance and provides decision-makers with multiple low-energy and high-profit disassembly schemes.

Identifying the cardinality-constrained critical nodes with a hybrid evolutionary algorithm

Identifying the critical nodes in a network is crucial for understanding its characteristics, controlling its structure, and determining its functionality. Cardinality-constrained CNP (CC-CNP) is a nondeterministic polynomial-time (NP)-hard combinatorial optimization problem that refers to minimizing a set of nodes such that after deletion, the size of the largest connected component in the residual subgraph is smaller than a prescribed value. CC-CNP is applicable to a variety of fields, such as epidemic and infectious disease control, electric power network construction and maintenance, and traffic network control.In this work, we present a multistage local search (MSLS) algorithm for generating high-quality initial solutions for CC-CNP, where two strategies, circular node deletion and node change and the tabu search-based first in, first out (FIFO) principle, are utilized to prevent search detours. Then, a population-based strategy is incorporated, resulting in a genetic algorithm-based multistage local search algorithm (GAMSLS) that adopts a genetic algorithm framework, refines the initial solutions in the crossover process, and utilizes a new population update strategy to ensure the diversity and individual quality of the population. The proposed algorithm is evaluated on 75 network instances and is shown to outperform state-of-the-art algorithms for CC-CNP.