Search

In this paper, a new technique to determine the best values of a PID controller is presented. The proposed scheme is based on using a single-neuron controller which its weights represent the PID parameters. Weight's adjustment is accomplished with a recent meta-heuristic algorithm called the DragonFly Algorithm. To show the effectiveness of our method, we have applied it to control a Continuous Stirred Tank Reactor. The obtained results are compared with several algorithms: the Ziegler–Nichols, Genetic Algorithm, and Particle Swarm Optimization. [ABSTRACT FROM AUTHOR]

Summary The channel availability problem reaches a higher degree in mobile ad hoc networks (MANETs) and garners a lot of attention in communication networks. Because increased mobile usage might result in a lack of channel allocation, an improved channel allocation technique is presented to tackle the availability problem. The distributed dynamic channel allocation (DDCA) model is built in this paper using the hybrid memory dragonfly with imperialist competitive (HMDIC) method. Based on optimization logic, this strategy assigns the channel to mobile hosts. The MANET provides a dispersed network within the coverage region in the absence of base station infrastructure. The HMDIC optimizer approach in this circumstance randomly begins every respective node to update and store their pbest value utilizing RAM dragonfly employing satellite images. The constraint values are then used to construct the cost function, which results in a strong kind of global optimum solution. The channels are therefore distributed in an effective manner. The HMDIC algorithm is used in this research to build a novel channel allocation system. It makes advantage of the exploration capabilities to successfully explore the individual node using MDA (Modified Dragonfly Algorithm) and locate the global best solution using imperialist competitive algorithm (ICA). Both of these combined tactics are more effective in accelerating the convergence of the allocation model. To validate the performance, the HMDIC‐based DDCA system provides promising results in terms of assigning available channels, thereby enhancing channel reuse efficiency and fractional interference. [ABSTRACT FROM AUTHOR]

The objective of this paper is to identify the most sensitive component of a turbogenerator and optimize its availability. To achieve this, we begin by conducting an initial reliability, availability, maintainability, and dependability (RAMD) analysis on each component. Subsequently, a novel stochastic model is developed to analyze the steady‐state availability of the turbogenerator, employing a Markov birth‐death process. In this model, failure and repair rates are assumed to follow an exponential distribution and are statistically independent. To optimize the proposed stochastic model, we employ four population‐based meta‐heuristic approaches: the grey wolf optimization (GWO), the dragonfly algorithm (DA), the grasshopper optimization algorithm (GOA), and the whale optimization algorithm (WOA). These algorithms are utilized to find the optimal solution by iteratively improving the availability of the turbogenerator. The performance of each algorithm is evaluated in terms of system availability and execution time, allowing us to identify the most efficient algorithm for this task. Based on the numerical results, it is evident that the WOA outperforms the GWO, GOA, and DA in terms of both system availability and execution time. [ABSTRACT FROM AUTHOR]

AbstractThis paper was conducted to model the drying kinetics of Capparis spinosa buds using dragonfly swarm optimized nonlinear regression and to compare the impact of drying treatments on their quality. Experiments included hot-air convective drying (HACD) from 40 to 120 °C, vacuum drying (VD) at 40, 60, and 80 °C, microwave drying (MD) ranging from 200 to 1000 W. Out of 30 models tested, the drying kinetics fitted best with Jena Das model in both HACD and VD and modified Midilli for MD. The comparison between the drying treatments used indicated that when taking into account the quality of the dried caper, VD at 80 °C was the most effective, resulting in a high-quality caper with a well-preserved color (ΔE = 7.47), high bioactives (TPC = 30.18mgGAE/gDW and TFC = 10.27 mg QE/gDW) and radical scavenging abilities (DPPH = 0.249 and ABTS = 4.448 mg/mL) at the expense of long duration (6h) and high specific energy consumption (50.667 kWh/kg). On the other hand, when considering the drying behavior MD ranked best with samples dried at 1000 W showing high diffusivities 2.04e−8 m2s−1 low energy consumption (MER 0.1216 kg/h), and short drying times (12 mn). This research is critical in selecting better drying conditions for increased usage in the culinary, cosmetic, and pharmaceutical sectors. [ABSTRACT FROM AUTHOR]

Knowledge of the optical properties of blood plays important role in medical diagnostics and therapeutic applications in laser medicine. In this paper, we present a very rapid and accurate artificial intelligent approach using Dragonfly Algorithm/Support Vector Machine models to estimate the optical properties of blood, specifically the absorption coefficient, and the scattering coefficient using key parameters such as wavelength (nm), hematocrit percentage (%), and saturation of oxygen (%), in building very highly accurate Dragonfly Algorithm-Support Vector Regression models (DA-SVR). 1000 training and testing sets were selected in the wavelength range of 250-1200 nm and the hematocrit of 0-100%. The performance of the proposed method is characterized by high accuracy indicated in the correlation coefficients (R) of 0.9994 and 0.9957 for absorption and scattering coefficients, respectively. In addition, the root mean squared error values (RMSE) of 0.972 and 2.9193, as well as low mean absolute error values (MAE) of 0.2173 and 0.2423, this result showed a strong match with the experimental data. The models can be used to accurately predict the absorption and scattering coefficients of blood, and provide a reliable reference for future studies on the optical properties of human blood. [ABSTRACT FROM AUTHOR]

Abstract Sediment phenomenon is very important in hydraulic and water resources issues. The existence of this phenomenon causes many problems in water storage. Sediment simulation in rivers helps in controlling sediment as well as reducing damages. In this study, an attempt was made to estimate the suspended sediment load using the corresponding river flow rate in the Zohreh River, Iran using the newest intelligent simulation methods. This study seeks to couple the nonlinear support vector regression (SVR) with crowd intelligence optimization algorithms. For this purpose, support vector regression was optimized using four new crowd optimization algorithms including the ant colony optimizer (ACO), the ant lion optimizer (ALO), the dragonfly algorithm (DA), and the salp swarm algorithm (SSA). Simulation was done in the two phases of train and test. Due to the integration of the nonlinear support vector regression with the optimization algorithms, the model train phase requires more time than usual situations. Therefore, in the current study, taking into account the number of different iterations including 25, 50, 100 and 200 iterations to perform the optimization of the model and tried to find the best optimizer by considering the calculated error and the run time. It was generally found that the SVR model is accurate in estimating the suspended sediment load. Finally, according to the calculated error as well as the run time, the support vector regression model optimized with the salp swarm algorithm with 25 iterations was chosen as the best model. Also, the values of R2, NSE, and RMSE for the best model in the test phase were calculated as 1, 1, and 10.2 tons per day, respectively, and the algorithm run time was 252 s.

In order to optimize the calibration set for near-infrared modeling of the protein content in wheat flour, the binary dragonfly algorithm (BDA) was used to select representative samples from the primary calibration set divided by the traditional Kennard/Stone (K/S) method. Based on the representative samples, a partial least square regression (PLSR) model for estimating the protein content in wheat flour was established, and the prediction set was employed to evaluate the stability and prediction performance of the model. The results indicated that an optimal calibration set with 30 samples was selected finally by BDA, and the proposed model exhibited a coefficient of determination of prediction (Rp2) of 0.956 4 and a root mean square errors of prediction (RMSEP) of 0.278 1, which increased by 1.87% and decreased by 15.57% compared with those (0.938 8 and 0.329 4) from K/S partition of 100 primary calibration sets, respectively. The average number of calibration sets selected from 10 BDA experiments was 30.2, and the protein content of wheat flour was predicted better by the 10 models developed than that obtained based on the primary calibration set. Therefore, BDA can select a small number of representative calibration set samples based on which a PLSR model with good robustness and high prediction accuracy for the protein content of wheat flour can be established. The proposed method can provide an efficient tool for calibration set selection in near-infrared spectroscopic analysis of the quality of wheat flour.

Swarm robotics show collective behavior in order to work in multi agent scenario, where spatial area coverage is an emergent area of research. This work introduces a hybrid technique for optimized spatial area coverage. A combination of Dragonfly Algorithm (DA) and chaotic mapping is proposed and the entire study is carried out in two phases. In the first phase, DA's parameters are measured on the basis of percentage of area covered, entropy and number of pop-up threats detected. In the second phase, Chaotic distribution is applied in swarm algorithms (i.e. DA along with Bat Algorithm and Accelerated Particle Swarm Optimization) to implement hybridized models. Chaotic Dragonfly Algorithm along with Chaotic Bat Algorithm and Chaotic Accelerated Particle Swarm Optimization) is implemented. To evaluate the performance of hybridized models, a comprehensive comparison is drawn among the Levy and Chaotic versions of all three swarm algorithms. It is concluded that the proposed DA outperforms the rest. DA not only showed better results for the mentioned metrics, but also displayed uniform behavior over multiple runs. [ABSTRACT FROM AUTHOR]

Wind energy forecasting studies play an important role in the search for sustainable energy solutions. However, wind power generation faces an inherent challenge. It is subject to constant fluctuations caused by meteorological conditions. These fluctuations can lead to inconsistencies in voltage and frequency within power grids, resulting in energy instability. To meet this challenge and ensure a reliable energy supply, measures must be taken to reduce the potential instability caused by changing wind conditions. This includes the development of advanced modeling techniques that take into account time-dependent and non-linear changes in wind speed. This type of modeling is crucial for minimizing energy losses and maintaining grid stability. As a result, the urgent need to meet the increasing energy demand while minimizing the environmental impact has triggered the transition to renewable energy sources. In this study, real short-term wind speed data from Osmaniye region were taken as research object. These data were analyzed in detail and the wind speed was estimated by considering the meteorological conditions. Artificial Neural Network was used in the prediction method, and the artificial intelligence algorithm was hybridized with the Dragonfly Algorithm and the coefficients of the artificial intelligence algorithm were trained with the Dragonfly Algorithm. It was used to compare the performance indexes of the prediction models designed with mean percent error, mean absolute percentage error, root mean square error. The performance analysis of Artificial Neural Network, Adaptive Neuro-Fuzzy Inference System, Fuzzy and Dragonfly-Based Artificial Neural Network are 2,2512,2,0698,1,7458 and 1,5212, respectively, based on mean absolute percentage error. Root mean square error values are 9,4857,8,2945,7,3285 and 6,4711. Finally, mean absolute errors are 8,2310, 7,5269, 6,2385 and 5,9486, respectively. [ABSTRACT FROM AUTHOR]

Pure zinc oxide (ZnO) thin films, along with manganese (Mn) doped counterparts, were produced using rapid thermal evaporation technique on ordinary glass substrates. Postannealing treatments resulted in the formation of hexagonal wurtzite structures in the deposited layers. The Raman results unveiled the presence of A1(LO) and LVM vibration modes in each sample that were doped. Interestingly, the undoped sample lacked the LVM mode while showcasing the emergence of LA + TO combined phonons. Employing a novel approach reliant on the Dragonfly Algorithm, optical parameters were extracted, revealing a drop in the bandgap energy of the films from 3.95 eV to 3.79 eV. [ABSTRACT FROM AUTHOR]

Copyright of Shipin Kexue/ Food Science is the property of Food Science Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Navigation poses a significant challenge for autonomous vehicles, prompting the exploration of various bio-inspired artificial intelligence techniques to address issues related to path generation, obstacle avoidance, and optimal path planning. Numerous studies have delved into bio-inspired approaches to navigate and overcome obstacles. In this paper, we introduce the dragonfly algorithm (DA), a novel bio-inspired meta-heuristic optimization technique to autonomously set goals, detect obstacles, and minimize human intervention. To enhance efficacy in unstructured environments, we propose and analyze the dragonfly–fuzzy hybrid algorithm, leveraging the strengths of both approaches. This hybrid controller amalgamates diverse features from different methods into a unified framework, offering a multifaceted solution. Through a comparative analysis of simulation and experimental results under varied environmental conditions, the hybrid dragonfly–fuzzy controller demonstrates superior performance in terms of time and path optimization compared to individual algorithms and traditional controllers. This research aims to contribute to the advancement of autonomous vehicle navigation through the innovative integration of bio-inspired meta-heuristic optimization techniques.

In view of the problems that the dragonfly algorithm has, such as that it easily falls into the local optimal solution and the optimization accuracy is low, an improved Dragonfly Algorithm (IDA) is proposed and applied to Otsu multi-threshold image segmentation. Firstly, an elite-opposition-based learning optimization is utilized to enhance the diversity of the initial population of dragonflies, laying the foundation for subsequent algorithm iterations. Secondly, an enhanced sine cosine strategy is introduced to prevent the algorithm from falling into local optima, thereby improving its ability to escape from local optima. Then, an adaptive t-distribution strategy is incorporated to enhance the balance between global exploration and local search, thereby improving the convergence speed of the algorithm. To evaluate the performance of this algorithm, we use eight international benchmark functions to test the performance of the IDA algorithm and compare it with the sparrow search algorithm (SSA), sine cosine algorithm (SCA) and dragonfly algorithm (DA). The experiments show that the algorithm performs better in terms of convergence speed and accuracy. At the same time, the Otsu method is employed to determine the optimal threshold, a series of experiments are carried out on six images provided by Berkeley University, and the results are compared with the other three algorithms. From the experimental results, the peak signal-to-noise ratio index (PSNR) and structural similarity index (SSIM) based on the IDA algorithm method are better than other optimization algorithms. The experimental results indicate that the application of Otsu multi-threshold segmentation based on the IDA algorithm is potential and meaningful. [ABSTRACT FROM AUTHOR]

In this paper, we present a novel DILTS algorithm that uses a new approach inspired by the energy efficiency of dragonflies. The algorithm optimizes the energy-harvesting mechanisms in IoT devices, inspired by the way dragonflies use wind energy to fly. A sophisticated algorithm optimizes power consumption during task execution, saving energy and speeding up tasks while maintaining the application throughput. The algorithm leverages lazy task scheduling (LTS) to enhance task execution performance. The proposed algorithm evaluates the energy levels of each task and implements an LTS method. This LTS approach improves performance and task management by streamlining scheduling data and reducing overhead. The LTS model reliably optimizes the energy across microbenchmarks and real-time IoT devices. To assess the efficiency and practicality of our algorithm, we compared it to four alternatives. Our novel algorithm outperformed the others with a chip area of 856 μm2, performance speed of 7.11 ns, scheduling accuracy of 94%, and response time of 2.61 ns. Our simulations showed that our proposed method reduced energy consumption by up to 10.02% compared to existing methods. We evaluated the performance of the algorithms on a Zynq 7000 FPGA using the Xilinx Vivado platform via simulations. Our novel algorithm can improve the energy efficiency of green data centers. [ABSTRACT FROM AUTHOR]

Navigation poses a significant challenge for autonomous vehicles, prompting the exploration of various bio-inspired artificial intelligence techniques to address issues related to path generation, obstacle avoidance, and optimal path planning. Numerous studies have delved into bio-inspired approaches to navigate and overcome obstacles. In this paper, we introduce the dragonfly algorithm (DA), a novel bio-inspired meta-heuristic optimization technique to autonomously set goals, detect obstacles, and minimize human intervention. To enhance efficacy in unstructured environments, we propose and analyze the dragonfly–fuzzy hybrid algorithm, leveraging the strengths of both approaches. This hybrid controller amalgamates diverse features from different methods into a unified framework, offering a multifaceted solution. Through a comparative analysis of simulation and experimental results under varied environmental conditions, the hybrid dragonfly–fuzzy controller demonstrates superior performance in terms of time and path optimization compared to individual algorithms and traditional controllers. This research aims to contribute to the advancement of autonomous vehicle navigation through the innovative integration of bio-inspired meta-heuristic optimization techniques. [ABSTRACT FROM AUTHOR]

Metaheuristics are commonly employed as a means of solving many distinct kinds of optimization problems. Several natural-process-inspired metaheuristic optimizers have been introduced in the recent years. The convergence, computational burden and statistical relevance of metaheuristics should be studied and compared for their potential use in future algorithm design and implementation. In this paper, eight different variants of dragonfly algorithm, i.e. classical dragonfly algorithm (DA), hybrid memory-based dragonfly algorithm with differential evolution (DADE), quantum-behaved and Gaussian mutational dragonfly algorithm (QGDA), memory-based hybrid dragonfly algorithm (MHDA), chaotic dragonfly algorithm (CDA), biogeography-based Mexican hat wavelet dragonfly algorithm (BMDA), hybrid Nelder-Mead algorithm and dragonfly algorithm (INMDA), and hybridization of dragonfly algorithm and artificial bee colony (HDA) are applied to solve four industrial chemical process optimization problems. A fuzzy multi-criteria decision making tool in the form of fuzzy-measurement alternatives and ranking according to compromise solution (MARCOS) is adopted to ascertain the relative rankings of the DA variants with respect to computational time, Friedman's rank based on optimal solutions and convergence rate. Based on the comprehensive testing of the algorithms, it is revealed that DADE, QGDA and classical DA are the top three DA variants in solving the industrial chemical process optimization problems under consideration. [ABSTRACT FROM AUTHOR]

Here we introduce a new evolutionary algorithm called the Lotus Effect Algorithm, which combines efficient operators from the dragonfly algorithm, such as the movement of dragonflies in flower pollination for exploration, with the self-cleaning feature of water on flower leaves known as the lotus effect, for extraction and local search operations. The authors compared this method to other improved versions of the dragonfly algorithm using standard benchmark functions, and it outperformed all other methods according to Fredman's test on 29 benchmark functions. The article also highlights the practical application of LEA in reducing energy consumption in IoT nodes through clustering, resulting in increased packet delivery ratio and network lifetime. Additionally, the performance of the proposed method was tested on real-world problems with multiple constraints, such as the welded beam design optimization problem and the speed-reducer problem applied in a gearbox, and the results showed that LEA performs better than other methods in terms of accuracy. [ABSTRACT FROM AUTHOR]

The development of the information industry is a cross-era impact for any industry, especially under the influence of the database, and the systematic analysis of the data in the industry can improve the accuracy of the output vector and data analysis. With the increase of data volume, for the difficulty of data processing increased significantly, the number of college students growing rapidly in recent years, to the accurate analysis of teachers and students need to take information technology, improve the accuracy of the operation, in the analysis of students and teachers through the system processing education data, the results can be applied to the education management. Based on the above content, on the basis of the dragonfly algorithm, through the improvement of the algorithm, the quality of rehabilitation physiology ideological teaching evaluation, first constructed the course ideological teaching quality evaluation system, then on the basis of determining the evaluation index, the teaching quality evaluation score, in order to through the study of rehabilitation physiology ideological teaching quality to provide reference. [ABSTRACT FROM AUTHOR]

Objective: In order to improve the precision of apple grade judgment model, the method of apple grade judgment was established. Methods: A decision model of apple rank based on multi-information fusion and dragonfly algorithm was proposed. Firstly, the HSV color feature, LBP texture feature and HOG shape feature of apple image were extracted by pre-processing such as data enhancement, normalization, Gauss filter and grayscale. Secondly, the performance of DBN model was affected by the selection of parameters, the network parameters of DBN model were optimized by DA algorithm, and a multi-information fusion and DA-DBN model for determining apple rank wws proposed. Results: Compared with GA-DBN, PSO-DBN, GWO-DBN and DBN, the model based on DA-DBN had the highest precision. Conclusion: The DBN model is optimized by dragonfly algorithm which can effectively improve the accuracy of apple rank determination model, which provides a new method for apple rank determination.

Brain-Computer Interfaces (BCIs) facilitate the translation of brain activity into actionable commands and act as a crucial link between the human brain and the external environment. Electroencephalography (EEG)-based BCIs, which focus on motor imagery, have emerged as an important area of study in this domain. They are used in neurorehabilitation, neuroprosthetics, and gaming, among other applications. Optimal Deep Learning-Based Recognition for EEG Signal Motor Imagery (ODLR-EEGSM) is a novel approach presented in this article that aims to improve the recognition of motor imagery from EEG signals. The proposed method includes several crucial stages to improve the precision and effectiveness of EEG-based motor imagery recognition. The pre-processing phase starts with the Variation Mode Decomposition (VMD) technique, which is used to improve EEG signals. The EEG signals are decomposed into different oscillatory modes by VMD, laying the groundwork for subsequent feature extraction. Feature extraction is a crucial component of the ODLR-EEGSM method. In this study, we use Stacked Sparse Auto Encoder (SSAE) models to identify significant patterns in the pre-processed EEG data. Our approach is based on the classification model using Deep Wavelet Neural Network (DWNN) optimized with Chaotic Dragonfly Algorithm (CDFA). CDFA optimizes the weight and bias values of the DWNN, significantly improving the classification accuracy of motor imagery. To evaluate the efficacy of the ODLR-EEGSM method, we use benchmark datasets to perform rigorous performance validation. The results show that our approach outperforms current methods in the classification of EEG motor imagery, confirming its promising performance. This study has the potential to make brain-computer interface applications in various fields more accurate and efficient, and pave the way for brain-controlled interactions with external systems and devices.

The deterministic directional overcurrent relay (DOCR) coordination approaches find limitation in providing fast and reliable protection to today's distribution networks due to growing integration of distributed generations (DGs). In this paper, an adaptive DOCR protection coordination scheme is proposed using an efficient optimization algorithm called the dragonfly algorithm (DA). The main inspiration of the proposed DA optimization technique originates from static and dynamic swarming behaviours of dragonflies. In this approach, the dominant changes in the network topologies are identified by monitoring the status of the circuit breakers connected at the terminals of the DGs and other main power components. When any dominant changes in the network topologies or operating modes are identified, the fault current for that particular condition is calculated and the new optimal DOCRs settings for the prevailing condition are obtained from the substation central computer in online mode. The comparative results with the existing approaches justify the superiority of the proposed scheme in achieving the minimum overall relay operating time and maintaining the coordination between the primary and backup relay pairs. In the proposed adaptive protection scheme, the average percentage reduction in the overall operating times of DOCRs in the 6-bus and the IEEE 14-bus test systems with different levels DG penetration is found to be 49.7% and 15%, respectively, compared to the existing approaches. [ABSTRACT FROM AUTHOR]

The state of health (SOH ) estimation of lithium batteries is low in accuracy while Gaussian process regression (GPR) is susceptible to local optimum and slow convergence speed. In this paper, the dragonfly algorithm (DA) is employed to optimize the hyperparameters of Gaussian process regression, and thus achieve accurate estimation of the nonlinear recession of lithium batteries. First, a number of features are selected from the charge/discharge cycle data of lithium batteries, including charging & discharging time and temperature change. Then, the correlation between these features and the SOH of lithium batteries is analyzed, and the features with higher correlation are selected as the health factors and taken as inputs to the model. The lithium battery degradation model is built by GPR and DA-GPR algorithms to predict the SOH of B0005, B0006 and B0007 on NASA battery datasets. The three sets of batteries are firstly charged in 1.5 A constant current mode (CC stage), and then changed to constant voltage mode when their voltage reaches the charging cut-off voltage of 4.2V (CV stage), and their charging is ended when the current of the batteries gradually decays to 20 mA. During the discharging process, the batteries are in 2A constant current mode until the battery B0005, B0006 and B0007 reaches the cut-off discharge voltage of 2.7 V, 2. 5 Vand 2. 2 V. Repeated charging and discharging will accelerate the aging process of lithium batteries. When the SOH of the test batteries is reduced to less than 70%, they reach their end of capacity and the experiment ends. The initial 75 cycles of B0005, B0006 and B0007 batteries are used as the training set and the remaining 90 cycles as the test set, in which the root mean square error (RMSE) for the Gaussian process regression (GPR) stands at 1.45, 0.285, and 2 768 4 respectively. In contrast, in the dragonfly algorithm-optimized Gaussian process regression (DA-GPR), the RMSE records at 0.119, 0.022 and 0.147, down by 91.79%, 92% and 94. 69% respectively. To verify the accuracy of DA-GPR prediction under limited number of samples, the first 10 cycles and the first 30 cycles of B0005 are selected as the training set, and the rest are used as the training set. Results show the RMSE of the first 10 cycles as the training set is 1.756 3, and that of the first 30 cycles as the training set is 0.150 71. To verify the generalizability of DA-GPR, B0006 and B0007 are chosen as training sets and B0005 as test set, whose RMSE reaches 0.74 4. In comparison with some other machine learning models, the RMSE of GPR reaches 1.545, of GA-BP (BP neural network optimized by genetic algorithm) 1 769 0, and of IGVO-SVR (support vector machine optimized by locust algorithm regression) 2.436. [ABSTRACT FROM AUTHOR]

For the successful operation of any industry or plant continuous availability of power supply is essential. Many of the large-scale plants established their power generation units. Marine power plant having two generators is also fall in this category. In this study, an effort is made to derive and optimize the availability of a marine power plant having two generators, one switch board and distribution switchboards. For this purpose, a mathematical model is proposed using Markov birth death process by considering exponentially distributed failure and repair rates of all the subsystems. The availability expression of marine power plant is derived. Metaheuristic algorithms namely dragonfly algorithm (DA), bat algorithm (BA) and whale optimization (WOA) are employed to optimize the availability of marine power plant. It is revealed that whale optimization algorithm outperforms over dragonfly algorithm (DA), and bat algorithm (BA) in optimum availability prediction and parameter estimation. The numerical values of the availability and estimated parameters are appended as numerical results. The derived results can be utilized in development of maintenance strategies of marine power plants and to carry out design modifications. [ABSTRACT FROM AUTHOR]

Esophageal squamous cell carcinoma (ESCC) is a type of cancer and has some of the highest rates of both incidence and mortality globally. Developing accurate models for survival prediction provides a basis clinical judgment and decision making, improving the survival status of ESCC patients. Although many predictive models have been developed, there is still lack of highly accurate survival prediction models for ESCC patients. This study proposes a novel survival prediction model for ESCC patients based on principal component analysis (PCA) and least-squares support vector machine (LSSVM) optimized by an improved dragonfly algorithm with hybrid strategy (HSIDA). The original 17 blood indicators are condensed into five new variables by PCA, reducing data dimensionality and redundancy. An improved dragonfly algorithm based on hybrid strategy is proposed, which addresses the limitations of dragonfly algorithm, such as slow convergence, low search accuracy and insufficient vitality of late search. The proposed HSIDA is used to optimize the regularization parameter and kernel parameter of LSSVM, improving the prediction accuracy of the model. The proposed model is validated on the dataset of 400 patients with ESCC in the clinical database of First Affiliated Hospital of Zhengzhou University and the State Key Laboratory of Esophageal Cancer Prevention and Control of Henan Province. The experiment results demonstrate that the proposed HSIDA-LSSVM has the best prediction performance than LSSVM, HSIDA-BP, IPSO-LSSVM, COA-LSSVM and IBA-LSSVM. The proposed model achieves the accuracy of 96.25%, sensitivity of 95.12%, specificity of 97.44%, precision of 97.50%, and F1 score of 96.30%. Graphical abstract [ABSTRACT FROM AUTHOR]

This study introduces novel stochastic distributed cooperative control (SDCC) in the context of island microgrids (MGs). A proportional resonant (PR) controller and virtual impedance droop control in stationary reference frames are employed in cooperation with distributed averaging secondary control optimized by the dragonfly algorithm (DA). The suggested approach demonstrates the capability to achieve mean-square synchronization for the voltage and frequency restoration of distributed generators (DGs) to ensure efficient active power sharing. Therefore, a sparse communication network has been used to avoid data congestion and reduce the need for extensive communication and information exchange. The proposed system offers an instinctive compromise between voltage regulation and reactive power sharing. A conventional centralized secondary control with PR droop control is simulated for performance evaluation and comparison purposes. In this study, empirical evidence is demonstrated to support the MG's ability to confront communication failure and its ability to work reliably during plug-and-play operations. [ABSTRACT FROM AUTHOR]

Concerns surrounding the environment along with inadequate energy supply and high cost of same are responsible for pollution, high energy demand, unpredictable and uneconomical power generation. These have contributed to the widespread agreement that sustainable renewable energy sources (RES) must be developed, particularly in isolated villages where expanding the grid may be challenging and financially unviable for power corporations. As a result, in order to effectively and cheaply utilize the plentiful renewable energy resources, an optimal sizing approach is required. This study is aimed at investigating the economic performance of the hybrid system of a stand-alone Small Hydropower/PV/Diesel generator with battery electricity production. The cost function was minimized using Dragonfly Algorithm (DA) in order to minimize the Cost of Energy (COE) generation. The decision variables are the number of small hydro turbines (NSHP), number of solar panels (NPV), number of batteries (NBATT) and the capacity of Diesel generator (PDG). The developed method is applied to a typical Kiri village in Shelleng Local Government area of Adamawa State. For uniformity, the hourly solar irradiance data were created by converting the monthly average solar irradiance data. A dragonfly optimization technique was utilized to reach an optimum solution for the hybrid system. The result obtained showed that the system components: small Hydropower, solar PV and Diesel generator were able to generate electrical power of 5,783,600 W, 56,259 W and 5.2941e-05 W respectively to meet the energy demand. Results obtained from the developed scheme were compared with those obtained when TORSCHE algorithm was used in optimizing the hybrid system. It was observed that a total energy cost of $5,224,500 was obtained for the developed technique while $5,839,600 was obtained as the total cost for the TORSCHE model. This showed that the developed scheme outperformed the system output from the TORSCHE algorithm in terms of cost of energy by 89.46%. [ABSTRACT FROM AUTHOR]

A wireless sensor network (WSN) consists of an extensive number of low-power sensor nodes to gather information from their environment and monitor physical activities. This makes node localization a crucial aspect in most WSN applications since measurement data is worthless unless the location from where the data is acquired is known precisely. The majority of localization solutions rely on anchor nodes for estimating the node locations with different localization accuracy, complexity, and hence different applicability. But, the cost and complexity in the localization of large-scale WSNs are not significantly reduced. In this paper, a novel Hybridized Dragonfly and Jaya Optimization technique (HyDAJ) is introduced for improving localization accuracy and performance of mobile WSNs. The proposed hybrid technique combines the advantages of Dragonfly algorithm and Jaya algorithm to localize the sensor nodes in a more efficient way and overcomes the limitations of the original algorithm. The hybrid algorithm verifies that all target nodes are precisely localized with higher accuracy. Simulation results reveal that HyDAJ outperforms existing methods under multiple metrics including localization efficiency, mean localization error, computation time, and energy consumption. [ABSTRACT FROM AUTHOR]

Cloud infrastructure provides a real time computing environment to customers and had wide applicability in healthcare, medical facilities, business, and several other areas. Most of the health data recorded and saved on the cloud. But the cloud infrastructure is configured using several components and that makes it a complex structure. And the high value of availability and reliability is essential for satisfactory operation of such systems. So, the present study is conducted with the prominent objective of assessing the optimum availability of the cloud infrastructure. For this purpose, a novel stochastic model is proposed and optimized using dragonfly algorithm (DA) and Grey Wolf optimization (GWO) algorithms. The Markovian approach is employed to develop the Chapman-Kolmogorov differential difference equations associate with the system. It is considered that all failure and repair rates are exponentially distributed. The repairs are perfect. The numerical results are derived to highlight the importance of the study and identify the best algorithm. The system attains its optimum availability 0.9998649 at population size 120 with iteration 700 by GWO. It is revealed that grey wolf optimization algorithm performed better than the Dragonfly algorithm in assessing the availability, best fitted parametric values and execution time. [ABSTRACT FROM AUTHOR]

Nowadays, many people are suffered from obesity, they tend to maintain their body weight by consuming a sufficient number of calories in their routine life. In this research, a novel Deep Learning-based Food Item Classification (DEEPFIC) approach has been proposed to categorize the different food items from the dataset with their calorie values. Initially, the images are processed using the sigmoid stretching method to enhance the image quality and remove the noises. Consequently, the pre-processed images are segmented using Improved Watershed Segmentation (IWS2) algorithm. RNN is used to extract features like shape, size, textures, and color. The extracted features are then normalized using the dragonfly technique. The Bi-LSTM is utilized to classify food products based on these pertinent aspects. The efficiency of the proposed method was calculated in terms of specificity, precision, accuracy, and recall F-measure. The proposed method improves the overall accuracy by 4.99%, 8.72%, and 10.4% better than the existing DCNN, FRCNN, and LSV-SVM methods respectively. [ABSTRACT FROM AUTHOR]

In recent years, WSNs are acquiring popularity due to small-sized and flexible implementation; many applications require quick data transfer with minimal energy consumption of nodes in the midst of the ubiquitous use of WSNs. These sensor nodes cover large regions according to application needs and choose the best optimal path. The main issue with WSN is how to cover the neighborhood correctly and send data to sink without falling into the trap of a single node and single route. Therefore, a recently researched approach namely the swarm-based dragonfly, which has been effectively used in miscellany applications is exploited for this work. The dragonfly method is based on the exploration phase using global search and exploitation phase using local search. The implicit swarming behaviors are thought to be the fundamental drive for routing algorithms. This paper introduce a Meta-heuristic based Optimized Opportunistic Routing Protocol for WSNs (MOORP) based upon the best optimal forwarder node selection and dragonfly route optimization. The forwarder node selection is optimized by residual energy and eucledian distance of the node. The path between forwarder and destination is identified by using the Dragon-fly algorithm. MOORP employs a route searching algorithm (RSA) and a Energy Level Matrix (ELM) update is used to enhancing the routing decision. The RSA finds an optimal path and selects the optimal forwarder node with the help of a heuristic update or ELM. MOORP performance is compared with other opportunistic routing protocols on important parameters such as the number of alive nodes, throughput, packet delivery ratio, message success rate, and average energy consumption,and also compare with pre-existing cluster based routing protocol. The simulation results show that the MOORP considerably outperforms its competitive techniques in terms of energy efficiency. [ABSTRACT FROM AUTHOR]

Unmanned aerial vehicles (UAVs) providing additional on-demand communication and computing services have become a promising technology. However, the limited energy supply of UAVs, which constrains their service duration, has emerged as an obstacle in UAV-enabled networks. In this context, a novel task offloading framework is proposed in UAV-enabled mobile edge computing (MEC) networks. Specifically, heterogeneous UAVs with different communication and computing capabilities are considered and the energy consumption of UAVs is minimized via jointly optimizing user association and UAV deployment. The optimal transport theory is introduced to analyze the user association sub-problem, and the UAV deployment for each sub-region is determined by a dragonfly algorithm (DA). Simulation results show that the energy consumption performance is significantly improved by the proposed algorithm. [ABSTRACT FROM AUTHOR]

Opportunistic routing is crucial for the development and management of an efficient and flexible network in Internet of Things (IoT) assisted wireless sensor network (WSN). Sensor nodes are dispersed over a large region and collect data from the surroundings. Data is analysed rather than sensed before being transferred via an established route to the base station (BS). Being a resource constrained network; overhead problems manifest. The energy consumption and subsequently, quick demise of nodes employed for data detection and transmission is major issue. Many other kind of network problems also arise in these constrained networks due to deadlock and livelock. Deadlock could occur if two nodes are trying to access the same data or resource simultaneously and neither can proceed until the other node releases it. This can lead to a complete halt in network operations, as nodes are unable to communicate or exchange data. The very purpose of deploying a network fails. Livelock could occur if two nodes are repeatedly sending messages to each other in response to previous messages, but none of the messages are leading to a resolution of the issue. This can lead to excessive use of network resources and can also cause a complete halt in network operations if the network becomes overloaded with traffic. This paper addresses the problem of deadlock and livelock in the context of IoT assisted WSN and proposed a novel approach based on the Adaptive meta-heuristic based Clustering and Routing Algorithm for IoT-assisted wireless sensor network (ACRA) to solve it. The primary contributions for the proposed method in such a context are: (1) Reduce Wastage of resources so that energy consumption is minimized (2) No single node stuck in the loop and thus removes hotspot problem (3) Use of alternate optimal path without getting stuck into a single path and finally, (4) the event must cover a large region. Two algorithms are proposed in ACRA including (1) the inter-deadlock avoidance clustering algorithm (INDCA) and (2) the intra-deadlock avoidance routing algorithm (INDRA). Both algorithms construct an optimal route and cover a large region. The proposed algorithm performance is compared with other similar algorithms on essential parameters such as the Packet Delivery Ratio (PDR), number of active nodes, and Average residual energy (ARE). The simulation results show that the proposed algorithm considerably outperforms its competitive techniques in terms of energy efficiency while at the same time addresses deadlock and livelock problems in a robust manner. [ABSTRACT FROM AUTHOR]

After the excavation of the roadway, the original stress balance is destroyed, resulting in the redistribution of stress and the formation of an excavation damaged zone (EDZ) around the roadway. The thickness of EDZ is the key basis for roadway stability discrimination and support structure design, and it is of great engineering significance to accurately predict the thickness of EDZ. Considering the advantages of machine learning (ML) in dealing with high-dimensional, nonlinear problems, a hybrid prediction model based on the random forest (RF) algorithm is developed in this paper. The model used the dragonfly algorithm (DA) to optimize two hyperparameters in RF, namely mtry and ntree, and used mean absolute error (MAE), root mean square error (RMSE), determination coefficient (R2), and variance accounted for (VAF) to evaluate model prediction performance. A database containing 217 sets of data was collected, with embedding depth (ED), drift span (DS), surrounding rock mass strength (RMS), joint index (JI) as input variables, and the excavation damaged zone thickness (EDZT) as output variable. In addition, four classic models, back propagation neural network (BPNN), extreme learning machine (ELM), radial basis function network (RBF), and RF were compared with the DA-RF model. The results showed that the DARF mold had the best prediction performance (training set: MAE = 0.1036, RMSE = 0.1514, R² = 0.9577, VAF = 94.2645; test set: MAE = 0.1115, RMSE = 0.1417, R² = 0.9423, VAF = 94.0836). The results of the sensitivity analysis showed that the relative importance of each input variable was DS, ED, RMS, and JI from low to high. [ABSTRACT FROM AUTHOR]