Iterative Genetic Algorithm Research Articles

Detector configuration optimization method considering response time for spatial source parameter estimation in complex urban wind environment

In source parameter estimation, improper arrangement of detectors may lead to ineffective concentration measurement or even failure in leakage detection, which would hold back emergency response efforts. However, current methods for optimizing detector configuration have inherent flaws and thus cannot be ideally applied in real-world scenarios. To address their limitations, this paper develops the method for optimizing detector configuration by considering response time. A novel cost function is proposed to balance response time and estimation accuracy based on specific demands. The optimum configuration is then obtained by genetic algorithm iteration. The wind tunnel test of an urban environment with buildings of varying heights is conducted for the research purpose. The superiority of the proposed method is demonstrated by comparing spatial source parameter estimations using optimum, uniform, and random configurations respectively. The results show that our method avoids placing detectors at locations with low concentrations, i.e., places where readings are invalid. It enables more detectors to obtain readings quickly, generating more accurate estimations in a shorter time. Only with the optimum configuration do detectors show readings for all 15 sources, with each unknown source having the highest number of detectors with readings. In contrast, the uniform configuration results in no readings for 4 sources and the random configuration results in no readings for 11 sources when leaking happens. The proposed method optimizes the detector configuration in three-dimensional space with complex flows, making it more feasible in practical applications.

HFRAS: design of a high-density feature representation model for effective augmentation of satellite images

AbstractEfficiently extracting features from satellite images is crucial for classification and post-processing activities. Many feature representation models have been created for this purpose. However, most of them either increase computational complexity or decrease classification efficiency. The proposed model in this paper initially collects a set of available satellite images and represents them via a hybrid of long short-term memory (LSTM) and gated recurrent unit (GRU) features. These features are processed via an iterative genetic algorithm, identifying optimal augmentation methods for the extracted feature sets. To analyse the efficiency of this optimization process, we model an iterative fitness function that assists in incrementally improving the classification process. The fitness function uses an accuracy & precision-based feedback mechanism, which helps in tuning the hyperparameters of the proposed LSTM & GRU feature extraction process. The suggested model used 100 k images, 60% allocated for training and 20% each designated for validation and testing purposes. The proposed model can increase classification precision by 16.1% and accuracy by 17.1% compared to conventional augmentation strategies. The model also showcased incremental accuracy enhancements for an increasing number of training image sets.

DLSDHMS: Design of a deep learning-based analysis model for secure and distributed hospital management using context-aware sidechains

Designing an efficient hospital management solution requires the integration of multidomain operations that include secure storage, alert system modelling, infrastructure management, staff management, report analysis, and feedback-based learning tasks. Existing hospital management models are either highly complex or do not incorporate comprehensive deep learning analysis, which limits their deployment capabilities. Moreover, most of these models use mutable storage solutions, which restricts their trust levels under multi-patient to multi-doctor mapping scenarios. To overcome these issues, this article proposes the design of a novel deep Learning-based analysis model for secure and distributed hospital management via context-aware sidechains. The model initially collects large-scale information sets from different hospital entities via an IoT-based network and stores the information on context-sensitive sidechains. These context-sensitive sidechains store information sets related to Medicine Management, Doctor Management, Insurance and Billing Management, and Appointment Management operations. These chains are optimized via an Iterative Genetic Algorithm (IGA) that assists in improving storage and retrieval performance via intelligent merging and splitting operations. Information stored on these chains is processed via a combination of Recurrent Neural Networks (RNNs) and Convolutional Neural Networks (CNNs), that assist in identifying patient-level diseases and issues. The information obtained from these classifiers is updated on the central repository and assists in the pre-emption of diseases for other patients. Due to these integrations, the proposed model is capable of reducing computational delay by 3.5 % and reducing storage cost by 8.3 % when compared to other blockchain-based deployments. The model is also able to pre-empt patient issues with 9.3 % higher accuracy and 4.8 % higher precision, which makes it useful for real-time clinical deployments.

Photogrammetry-based bending monitoring and load identification of steel truss structures

The bending of steel truss structures is an important gauge for detecting, identifying, and evaluating potential issues with structural safety performance. The limitations and high cost of traditional monitoring methods make it challenging to carry out stable long-term monitoring. Therefore, this paper developed a displacement monitoring system for steel truss structures which fulfill the requirements of having low cost, high stability, and ease of operation. The system is based on the improved sub-pixel positioning technology, achieving precise positioning in unfavorable conditions such as long structure-camera distance, angle skew, and dim light. Then, this system was calibrated through field experiment and compared with other measurement systems. Finally, a load identification method was developed to identify discrepancies between the true load and the design load. This method uses optimization functions to identify the true load applied in the experiment, and the optimization parameter obtained by a genetic algorithm iteration is output as the optimal solution. The results suggest that the photogrammetric system performs well in practical engineering applications and can provide advantages including high precision, low cost, simple operation, etc. Results obtained by the load identification method agree well with measurements obtained from the actual structure, and can serve as a tool for evaluating the mechanical properties of similar structures. This method monitors potential risks of steel truss structures, and greatly improve the stability and safety of such structures.

Optimal configuration for multi-purpose mine water quality-classified utilization

ABSTRACT Comprehensive utilization of mine water is of immense significance to avoid water wastage and protect the environment. To maximize the benefit of mine water utilization by calculating the available water in the goaf of Wucun Mine and analyzing the characteristics of water quality, combined with the water requirements of residents, power plants, irrigation, aquaculture, and ecology around the mining area and their water quality requirements, a mathematical model for optimal allocation of water quality-classified utilization in a 718-m3/h mine was constructed and solved using differential evolution algorithm and genetic algorithm, respectively. While the results of seven iterations of genetic algorithm varied between 215.24 and 232.36 China Yuan/h, those of differential evolution algorithm were all 232.36 China Yuan/h. When compared with genetic algorithm, differential evolution algorithm has the advantages of better convergence and stable and reliable calculation results. In the irrigation period, differential evolution algorithm indicated supply of 6.00, 334.95, 353.89, 19.76, and 3.40 m3/h treated mine water for residential usage, power plant, irrigation, aquaculture, and ecology, respectively, whereas in the non-irrigation period, differential evolution algorithm suggested supply of 6.00, 688.84, 0, 19.76, and 3.40 m3/h treated mine water, respectively. The annual income from quality-classified utilization of the mine water was the highest. Thus, the present study provides a model that can be used for reference for rational utilization of mine water resources.

Optimization design of steady optical pin beam using genetic algorithm

In this letter, an optimal designed steady optical beam (SOB) based on genetic algorithm (GA) is proposed. In the method, the parameters of the SOB are mapped in a novel way, and continuously optimized in the genetic algorithm iteration. The influence of these parameters on SOB propagation behavior is analyzed and explored by multiple iterations. In both simulation and experiment, the optimized SOBs are proved to exhibit better propagation features with smaller beam waist radius and divergence than the non-optimized ones. In the experiment, the beam parameter product (BPP) of the optimized SOB reaches 67.0% of the non-optimized SOB within a certain distance range. Our method is proved to have good performance and may find broad application prospects in the fields of optical communication, biomedicine, particle capture, material processing and so on.

Model-Assisted Online Optimization of Gain-Scheduled PID Control Using NSGA-II Iterative Genetic Algorithm

In the practical control of nonlinear valve systems, PID control, as a model-free method, continues to play a crucial role thanks to its simple structure and performance-oriented tuning process. To improve the control performance, advanced gain-scheduling methods are used to schedule the PID control gains based on the operating conditions and/or tracking error. However, determining the scheduled gain is a major challenge, as PID control gains need to be determined at each operating condition. In this paper, a model-assisted online optimization method is proposed based on the modified Non-Dominated Sorting Genetic Algorithms-II (NSGA-II) to obtain the optimal gain-scheduled PID controller. Model-assisted offline optimization through computer-in-the-loop simulation provides the initial scheduled gains for an online algorithm, which then uses the iterative NSGA-II algorithm to automatically schedule and tune PID gains by online searching of the parameter space. As a summary, the proposed approach presents a PID controller optimized through both model-assisted learning based on prior model knowledge and model-free online learning. The proposed approach is demonstrated in the case of a nonlinear valve system able to obtain optimal PID control gains with a given scheduled gain structure. The performance improvement of the optimized gain-scheduled PID control is demonstrated by comparing it with fixed-gain controllers under multiple operating conditions.

Optimal Partition Assignment for Universal Object Detection

The label assignment problem is a core task in object detection, which mainly focuses on how to define the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$positive/negative$</tex-math></inline-formula> samples during the training phase. Recent works have proved that label assignment is significant for performance improvement of the detector. In this paper, we propose an exquisite strategy that can dynamically assign labels according samples' joint scores (classification and location). Moreover, our strategy can apply to both 2D and 3D monocular detectors. In our strategy, we formulate label assignment as an optimization problem. Concretely, we first calculate the classification and location costs of each sample, which are treated as points in a two-dimensional coordinate system. Then an optimal divider line that minimizes the sum of point-to-line distances is designed to separate the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$positive/negative$</tex-math></inline-formula> samples. An iterative Genetic Algorithm is employed in acquiring the optimal solution. Furthermore, a GIoU auxiliary branch is devised to keep sample selection consistent during the training and testing phase. Benefitting from the non-maximum suppression (NMS) that utilizes the joint scores of classification and location, excellent detection performance is achieved. Extensive experiments conducted on MS COCO, PASCAL VOC (2D object detection), and KITTI (3D object detection) verify the effectiveness and universality of our proposed Optimal Partition Assignment (OPA).

Robust Heading Estimation Algorithm for Android Smartphones

This article proposes a robust heading estimation algorithm for smartphones, which integrates the gradient descent (GD) approach and genetic algorithm (GA) to compute the heading angle of smartphones and reduce the heading error accumulation induced by the errors of sensors. In the proposed algorithm, the accelerometer, magnetometer, and gyroscope are used to estimate the heading angle. The sum of squared triaxial errors of geomagnetic strength and acceleration is regarded as the fitness of GA to improve the accuracy of heading angle and restrain the error accumulation. The angular velocity given by gyroscope is used to generate the init ial point of GA. To improve the convergence rate and local search capability of GA, a search area adjustment strategy based on GD is designed to obtain the optimal search area for each iteration of GA. Experimental results show that the proposed algorithm can achieve better heading estimation accuracy on three smartphones and four orientations, with a mean heading error of 1.577° on Mate 20 smartphone.

Robust job-shop scheduling under deterministic and stochastic unavailability constraints due to preventive and corrective maintenance

This paper presents a modeling and solution approach for a robust job-shop scheduling problem (JSP) under deterministic and stochastic machine unavailability caused, respectively, by planned preventive maintenance (PM) and unplanned corrective maintenance (CM) following random breakdowns. The goal is to optimize the sequence of jobs and run-based planned preventive maintenance tasks in a robust manner while considering the degradation of machines over time. The time to failure for each machine is assumed to follow a Weibull distribution. The robust objective function to be minimized is a weighted sum of the expected values of the makespan and the gross positive deviation between the actual and planned start times of the jobs, as proxies for quality robustness and solution robustness, respectively. Two metaheuristic algorithms that aim at providing a good balance between performance quality and solution robustness are developed. In both algorithms, the “true” makespan objective function (with both PM and CM) is approximated using three surrogate measures. Genetic algorithm (GA) is first used to optimize the surrogate functions, then the fittest solutions from the three oracles are simulated with random breakdown scenarios and the best among them is introduced as an elite member in the next GA iteration. The first algorithm terminates as soon as the solution obtained is worse than the best know solution or when the maximum number of iterations is reached, whereas the second algorithm applies a rule inspired by Simulated Annealing for termination. Numerical experimentation on benchmark instances from the literature showed excellent performance of the proposed algorithms in terms of average marginal improvement and runtime. These results show that the proposed framework can generate high-quality robust job shop schedules under deterministic and stochastic machine unavailability constraints. Moreover, the sensitivity analysis results recommended some key insights and directions for future research.

EEG electrodes selection for emotion recognition independent of stimulus presentation paradigms

Most of the electroencephalography (EEG) based emotion recognition systems rely on single stimulus to evoke emotions. EEG data is mostly recorded with higher number of electrodes that can lead to data redundancy and longer experimental setup time. The question “whether the configuration with lesser number of electrodes is common amongst different stimuli presentation paradigms” remains unanswered. There are publicly available datasets for EEG based human emotional states recognition. Since this work is focused towards classifying emotions while subjects are experiencing different stimuli, therefore we need to perform new experiments. Keeping aforementioned issues in consideration, this work presents a novel experimental study that records EEG data for three different human emotional states evoked with four different stimuli presentation paradigms. A methodology based on iterative Genetic Algorithm in combination with majority voting has been used to achieve configuration with reduced number of EEG electrodes keeping in consideration minimum loss of classification accuracy. The results obtained are comparable with recent studies. Stimulus independent configurations with lesser number of electrodes lead towards low computational complexity as well as reduced set up time for future EEG based smart systems for emotions recognition

Spectrometric prediction of wood basic density by comparison of different grain angles and variable selection methods

BackgroundWood basic density (WBD) is one of the most crucial wood property in tree and mainly determined the end use of wood for industry. However, the measurement WBD is time- and cost-consuming, which an alternatively fast and no-destructive measurement is needed. In this study, capability of NIR spectroscopy combined with partial least squares regression (PLSR) to quantify the WBD were examined in multiple wood species. To obtain more accurate and robust prediction models, the grain angle (0° (transverse surface), 45°, 90° (radial surface)) influence on the collection of solid wood spectra and a comparison of found variable selection methods for NIR spectral variables optimization were conducted, including significant Multivariate Correlation (sMC), Regularized elimination procedure (Rep), Iterative predictor weighting (Ipw) and Genetic algorithm (Ga). Models made by random calibration data selection were conducted 200 times performance evaluation.ResultsThese results indicate that 90° angle models display relatively highest efficiency than other angle models, mixed angle model yield a satisfied WBD prediction results as well and could reduce the influence of grain angle. Rep method shows a higher efficiency than other methods which could eliminate the uninformative variables and enhance the predictive performance of 90° angle and mix angle models.ConclusionsThis study is potentially shown that the WBD (g/cm3) on solid wood across grain angles and varies wood species could be measured in a rapid and efficient way using NIR technology. Combined with the PLSR model, our methodology could serve as a tool for wood properties breeding and silviculture study.

Proposal of Iterative Genetic Algorithm for Test Suite Generation

Huge and reputed software industries are expected to deliver quality products. However, industry suffers from a loss of approximately $500 billion due to shoddy software quality. The quality of the product in terms of its accuracy, efficiency, and reliability can be revamped through testing by focusing attention on testing the product through effective test case generation and prioritization. The authors have proposed a test-case generation technique based on iterative listener genetic algorithm that generates test cases automatically. The proposed technique uses its adaptive nature and solves the issues like redundant test cases, inefficient test coverage percentage, high execution time, and increased computation complexity by maintaining the diversity of the population which will decrease the redundancy in test cases. The performance of the technique is compared with four existing test-case generation algorithms in terms of computational complexity, execution time, coverage, and it is observed that the proposed technique outperformed.

Effective PCB Decoupling Optimization by Combining an Iterative Genetic Algorithm and Machine Learning

An iterative optimization for decoupling capacitor placement on a power delivery network (PDN) is presented based on Genetic Algorithm (GA) and Artificial Neural Network (ANN). The ANN is first trained by an appropriate set of results obtained by a commercial simulator. Once the ANN is ready, it is used within an iterative GA process to place a minimum number of decoupling capacitors for minimizing the differences between the input impedance at one or more location, and the required target impedance. The combined GA–ANN process is shown to effectively provide results consistent with those obtained by a longer optimization based on commercial simulators. With the new approach the accuracy of the results remains at the same level, but the computational time is reduced by at least 30 times. Two test cases have been considered for validating the proposed approach, with the second one also being compared by experimental measurements.

Phase-controlled metasurface design via optimized genetic algorithm

Abstract In an optical Pancharatnam-Berry (PB) phase metasurface, each sub-wavelength dielectric structure of varied spatial orientation can be treated as a point source with the same amplitude yet varied relative phase. In this work, we introduce an optimized genetic algorithm (GA) method for the synthesis of one-dimensional (1D) PB phase-controlled dielectric metasurfaces by seeking for optimized phase profile solutions, which differs from previously reported amplitude-controlled GA method only applicable to generate transverse optical modes with plasmonic metasurfaces. The GA–optimized phase profiles can be readily used to construct dielectric metasurfaces with improved functionalities. The loop of phase-controlled GA consists of initialization, random mutation, screened evolution, and duplication. Here random mutation is realized by changing the phase of each unit cell, and this process should be efficient to obtain enough mutations to drive the whole GA process under supervision of appropriate mutation boundary. A well-chosen fitness function ensures the right direction of screened evolution, and the duplication process guarantees an equilibrated number of generated light patterns. Importantly, we optimize the GA loop by introducing a multi-step hierarchical mutation process to break local optimum limits. We demonstrate the validity of our optimized GA method by generating longitudinal optical modes (i. e., non-diffractive light sheets) with 1D PB phase dielectric metasurfaces having non-analytical counter-intuitive phase profiles. The produced large-area, long-distance light sheets could be used for realizing high-speed, low-noise light-sheet microscopy. Additionally, a simplified 3D light pattern generated by a 2D PB phase metasurface further reveals the potential of our optimized GA method for manipulating truly 3D light fields.

Greedy iterative genetic algorithm for the volume-based cross dock transportation problem associated with fixed charge

Cross docking is a distribution method wherein the products from inbound vehicles are loaded directly onto outbound vehicles with a minor or no storage in between. The fixed charge cross dock transportation problem (FCCDTP) involves determining the optimal method of loading and routing the vehicles in a cross dock system with minimal fixed and variable cost of transportation. In this paper, the FCCDTP with the volume-based quantification of products is formulated as a mixed integer linear programming model. The model is solved using the optimisation software LINGO solver, the proposed GA with variable neighbourhood search (GA-VNS) meta-heuristic and the proposed greedy iterative genetic algorithm (GIGA) meta-heuristic. The data for the problems are obtained from a real-life logistics company. The analysis of results reveals that the proposed GIGA meta-heuristic provides lesser total transportation cost as compared to the GA-VNS meta-heuristic.

Greedy iterative genetic algorithm for the volume based cross dock transportation problem associated with fixed charge

Cross docking is a distribution method wherein the products from inbound vehicles are loaded directly onto outbound vehicles with a minor or no storage in between. The fixed charge cross dock transportation problem (FCCDTP) involves determining the optimal method of loading and routing the vehicles in a cross dock system with minimal fixed and variable cost of transportation. In this paper, the FCCDTP with the volume-based quantification of products is formulated as a mixed integer linear programming model. The model is solved using the optimisation software LINGO solver, the proposed GA with variable neighbourhood search (GA-VNS) meta-heuristic and the proposed greedy iterative genetic algorithm (GIGA) meta-heuristic. The data for the problems are obtained from a real-life logistics company. The analysis of results reveals that the proposed GIGA meta-heuristic provides lesser total transportation cost as compared to the GA-VNS meta-heuristic.

Study and Optimal Design of a Direct-Driven Stator Coreless Axial Flux Permanent Magnet Synchronous Generator with Improved Dynamic Performance

In this paper, the study and optimization design of stator coreless axial flux permanent magnet synchronous generators is presented for direct driven variable speed renewable energy generation system applications while considering the requirement of reliability and dynamic performance with unstable input conditions. The dynamic analytical model is developed based on the investigation of the axial flux permanent magnet synchronous generator (AFPMSG) structure and basic electromagnetic equations to find out the relationship between generator parameters and dynamic performance. Simulation via the MATLAB/Simulink platform is carried out to obtain the sensitivity of dynamic performance to generator parameters. An integrated optimization model that takes the key parameters as variables is proposed, aiming to improve the mechanical dynamic performance of AFPMSG. For accurate design, the design procedure is modified by combining the nonlinear iterative genetic algorithm (GA) to perform the calculation. A 3_kW AFPMSG is optimally designed to minimize the output voltage overshooting—the index of dynamic performance for direct driven variable speed generation application. Finally, a three-dimensional (3D) finite element model of the generator is established by Maxwell ANSOFT, and the simulation results confirm the validity of the dynamic performance analysis and optimal design procedure.

Calculation of Detector Positions for a Source Localizing Radiation Portal Monitor System Using a Modified Iterative Genetic Algorithm

Calculation of Detector Positions for a Source Localizing Radiation Portal Monitor System Using a Modified Iterative Genetic Algorithm

An innovative framework for designing genetic algorithm structures

Abstract Genetic Algorithms are popular optimization algorithms, often used to solve complex large scale optimization problems in many fields. Like other meta-heuristic algorithms, Genetic Algorithms can only provide a probabilistic guarantee of the global optimal solution. Having a Genetic Algorithm (GA) capable of finding the global optimal solution with high success probability is always desirable. In this article, an innovative framework for designing an effective GA structure that can enhance the GA's success probability of finding the global optimal solution is proposed. The GA designed with the proposed framework has three innovations. First, the GA is capable of restarting its search process, based on adaptive condition, to jump out of local optima, if being trapped, to enhance the GA's exploration. Second, the GA has a local solution generation module which is integrated in the GA loop to enhance the GA's exploitation. Third, a systematic method based on Taguchi Experimental Design is proposed to tune the GA parameter set to balance the exploration and exploitation to enhance the GA capability of finding the global optimal solution. Effectiveness of the proposed framework is validated in 20 large-scale case study problems in which the GA designed by the proposed framework always outperforms five other algorithms available in the global optimization literature.