Maximum Iteration Research Articles

Performance improvement of the solid oxide fuel cell using optimal parameters identification through fuzzy logic based-modeling and different optimization algorithms

Fuel Cells are novel devices that have been proposed as new power generation systems. The advantages of solid oxide fuel cells are higher efficiency, higher stability, fuel flexibility, lower emissions, and generally lower cost. In the present study, the fuzzy model is employed to build the model of the solid oxide fuel cell considering various sputtering power, thickness of electrolyte, and temperatures of cell. The maximum iterations for the adaptive neuro-fuzzy inference model was considered 50 iterations. About 3500 samples were applied for the training process, and almost 900 samples were utilized for the testing. After modeling process, the genetic algorithm, particle swarm, simulated annealing, and hybrid firefly-particle swarm optimizers are applied to achieve the optimum value of current and power densities. The results showed that proposed fuzzy model could approximate the model the system with a good agreement with experimental data. Additionally, the obtained data confirm the accuracy, high convergence speed, and robustness of the proposed hybrid optimizer compared to three efficient optimization algorithms. Accordingly, the correlation factor for the proposed fuzzy model for the training and testing dataset was obtained to be 0.9298 and 0.9289, correspondingly.

Application of hybrid teaching mode in translation teaching from the perspective of ecological concept teaching

Blended teaching is a model of teaching that combines traditional classroom teaching with online teaching. The integration of blended teaching is not only the integration of carrier means but also the breakthrough of teaching mode and the expansion of teaching time and space. Based on the theory of education ecology, this paper first analyzes the mixture of college English translation teaching modes and indicates that the college English translation teaching system from the dimension of the language training system, cultivation system, and communication dimensions cultural dimensions of three parts, and communicative dimension culture contains three platforms, four interaction, five links. Then, the hardware and software of the hybrid teaching ecosystem are designed to realize translation teaching. In the test to verify the system performance in this paper, it was found that the overall iteration value of the system presented a rising trend with the increasing number of tests, and in 6 tests, the maximum iteration value of the system in this paper was always 3-12 higher than the maximum iteration value of the teaching system in the traditional protocol data test group. Therefore, the hybrid teaching ecosystem designed in this paper has obvious advantages in reliability and practicality.

OPTIMIZATION OF K-MEANS CLUSTERING USING PARTICLE SWARM OPTIMIZATION ALGORITHM FOR GROUPING TRAVELER REVIEWS DATA ON TRIPADVISOR SITES

K-Means Algorithm can be used to group tourists based on reviews on tourist destination objects. This algorithm has a weakness that is sensitive to the determination of the initial centroid. The initial centroid that is determined at random will decreasing the level accuracy, often gets stuck at the local optimum, and gets a random solution. Optimization algorithms such as PSO can overcome this by determining the optimal initial centroid. The optimal number of clusters (K) will be determined using the Elbow method by calculating the SSE value of the resulting cluster. The average Silhouette Coefficient (SC) is used to measure the quality of the clusters produced by the K-Means Algorithm with and without the PSO Algorithm. This study uses secondary data obtained from the UCI Machine Learning Repository with the name Travel Reviews Data Set which consists of 980 records and 10 attributes. The test results show that K=2 is the optimal number of clusters. The K-Means and PSO Algorithm gives an average SC value of 0.300358 which is better than without the PSO Algorithm of 0.300076. The optimal PSO hyperparameter generated is the number of particles=30, \varphi_1=2.2, and {\ \varphi}_2=3 at maximum iteration of 100.

A High-Efficient Reinforcement Learning Approach for Dexterous Manipulation.

Robotic hands have the potential to perform complex tasks in unstructured environments owing to their bionic design, inspired by the most agile biological hand. However, the modeling, planning and control of dexterous hands remain unresolved, open challenges, resulting in the simple movements and relatively clumsy motions of current robotic end effectors. This paper proposed a dynamic model based on generative adversarial architecture to learn the state mode of the dexterous hand, reducing the model's prediction error in long spans. An adaptive trajectory planning kernel was also developed to generate High-Value Area Trajectory (HVAT) data according to the control task and dynamic model, with adaptive trajectory adjustment achieved by changing the Levenberg-Marquardt (LM) coefficient and the linear searching coefficient. Furthermore, an improved Soft Actor-Critic (SAC) algorithm is designed by combining maximum entropy value iteration and HVAT value iteration. An experimental platform and simulation program were built to verify the proposed method with two manipulating tasks. The experimental results indicate that the proposed dexterous hand reinforcement learning algorithm has better training efficiency and requires fewer training samples to achieve quite satisfactory learning and control performance.

Leopard seal optimization (LSO): A natural inspired meta-heuristic algorithm

The main objective of this paper is to introduce a new NIO algorithm inspired from the hunting strategy of the leopard seals called Leopard Seal Optimization (LSO)to provide a simple swarm intelligence algorithm that have a high flexibility to solve real-time engineering problems in a fast and more accurate manner without falling into local optima problem. LSO is compared against recent NIO algorithms considering feature selection for disease diagnosing as the underlying optimization problem. In experimental results, LSO has been statistically tested against a recent six swarm intelligence algorithms using Wilcoxon test and t-test with significance level equals 0.05 based on the common five benchmark functions. These recent algorithms are called Tuna Swarm Optimization (TSO), Pelican Optimization Algorithm (POA), Cat and Mouse-Based Optimization (CMBO), Aphid–Ant Mutualism (AAM), White Shark Optimizer (WSO), and Red Piranha Optimization (RPO). The statistical analysis proved that LSO outperforms other recent techniques in most cases where the most tested results are less than 0.05. Then, LSO has been tested against the same algorithms in binary version as feature selection algorithms using two metrics called accuracy and execution time. It is noted that LSO is faster and more accurate than other algorithms to determine the optimal set of features at all numbers of search agents. Finally, it is concluded that LSO outperforms other techniques using accuracy, execution time, Wilcoxon test, and t-test metrics. Compared to the recent techniques using the maximum iteration number=400, LSO provided the maximum accuracy value that equals 97.62% at the search agents number=120 and the minimum execution time that equals 1314 at the search agents number= 40.

Design and Microscale Fabrication of Negative Poisson’s Ratio Lattice Structure Based on Multi-Scale Topology Optimization

The current design of negative Poisson’s ratio lattice structures is mainly forward-looking and predominantly dependent on several known deformation patterns. To automate the generation of structures with programmable Poisson’s ratio, the study utilized the energy homogenization method and the Solid Isotropic Material with Penalization (SIMP) method to establish an optimization model for negative Poisson’s ratio. By proposing a relaxed objective function and eliminating damping in the Optimality Criteria (OC) method, the study achieves the automatic evolution of negative Poisson’s ratio programmable lattice unit cells, with the lowest Poisson’s ratio achieving −0.5367, and an equivalent elastic matrix is derived. The iterative process’s efficiency is comparable to that of commercial software, with a maximum iteration time of 300 s, enabling the prompt identification of fundamental configurations. To validate the method’s effectiveness, finite element analysis was performed on four tubular structures, revealing evident tension–compression deformation patterns. Moreover, the microscale selective laser melting was used to successfully prepare multiple sets of tubular samples made from 316L stainless steel, each with a height of 5 mm. Quasi-static compression experiments showed negative Poisson’s ratio effects and buckling forms that align with finite element analysis results, providing valuable insights for industry applications.

Wrap-Around Based Frequency-Varying Viterbi Detection for Windowed OFDM Systems in Doubly Selective Fading Channels

In doubly selective fading channels, the conventional orthogonal frequency division multiplexing (OFDM) system suffers high inter-carrier interferences (ICIs). To deal with the ICIs, combining the time-domain windowing and Viterbi detection, we propose a wrap-around based frequency-varying Viterbi detection scheme for windowed OFDM systems. In a doubly selective fading channel, we theoretically derive the input-output relationship of windowed OFDM systems. It is observed that by selecting a proper window function, the equivalent channel matrix is sparse. Therefore, the severe ICIs can be significantly mitigated by our proposed wrap-around based frequency-varying Viterbi detection scheme. It is shown by simulation results that our proposed scheme achieves the performance very close to that of the maximum likelihood detector, even when the normalized Doppler frequency is as high as 0.667. Furthermore, the computational complexity of our proposed detection method is linear to the number of subcarriers, given the maximum iteration times and the constraint length.

A novel algorithm for solving sum of several affine fractional functions

<abstract><p>By using the outer space branch-and-reduction scheme, we present a novel algorithm for globally optimizing the sum of several affine fractional functions problem (SAFFP) over a nonempty compact set. For providing the reliable lower bounds in the searching process of iterations, we devise a novel linearizing method to establish the affine relaxation problem (ARP) for the SAFFP. Thus, the main computational work involves solving a series of ARP. For improving the convergence speed of the algorithm, an outer space region reduction technique is proposed by utilizing the objective function characteristics. Through computational complexity analysis, we estimate the algorithmic maximum iteration times. Finally, numerical comparison results are given to reveal the algorithmic computational advantages.</p></abstract>

Research on Coal Dust Wettability Identification Based on GA-BP Model.

Aiming at the problems of the influencing factors of coal mine dust wettability not being clear and the identification process being complicated, this study proposed a coal mine dust wettability identification method based on a back propagation (BP) neural network optimized by a genetic algorithm (GA). Firstly, 13 parameters of the physical and chemical properties of coal dust, which affect the wettability of coal dust, were determined, and on this basis, the initial weight and threshold of the BP neural network were optimized by combining the parallelism and robustness of the genetic algorithm, etc., and an adaptive GA−BP model, which could reasonably identify the wettability of coal dust was constructed. The extreme learning machine (ELM) algorithm is a single hidden layer neural network, and the training speed is faster than traditional neural networks. The particle swarm optimization (PSO) algorithm optimizes the weight and threshold of the ELM, so PSO−ELM could also realize the identification of coal dust wettability. The results showed that by comparing the four different models, the accuracy of coal dust wettability identification was ranked as GA−BP > PSO−ELM > ELM > BP. When the maximum iteration times and population size of the PSO algorithm and the GA algorithm were the same, the running time of the different models was also different, and the time consumption was ranked as ELM < BP < PSO−ELM < GA−BP. The GA−BP model had the highest discrimination accuracy for coal mine dust wettability with an accuracy of 96.6%. This study enriched the theory and method of coal mine dust wettability identification and has important significance for the efficient prevention and control of coal mine dust as well as occupational safety and health development.

Implementation of The Backpropagation Method and The Kohonen Network to Predict Blood Availability: Case Study in PMI Kota Magelang

The availability of blood stocks at the Indonesian Red Cross or Palang Merah Indonesia (PMI) is a must and absolute for institutions that organize the procurement and distribution of blood for medical purposes. The problem is that the blood stock in PMI Magelang City Branch is not ideally available in each blood type, especially blood type AB, which in recent years has been very minimal and difficult to obtain. The purpose of this study was to predict blood stock of type AB with software based on artificial neural networks backpropagation and Coherent tissues. Artificial Neural Network (JST) backpropagation is used to predict the stock supply of blood type AB. Meanwhile, Coherent is a network used to divide input patterns into groups. The application has a network structure consisting of 2 input neurons, 10 neurons in the hidden layer, and 1 and 1 neuron in the output layer. The total amount of data is 3 years (2015-2017), 2 years of data are used for training data, and 1 year of data is used for testing data. The engine predicts using a maximum iteration of 1,000 epochs, an expletive constant of 0.5, a momentum of 0.9, and a minimum error rate of 0.001. With variations in the value of the backpropagation component, a prediction of less than 140 bags per year is generated. Meanwhile, the resulting weight is predicted by the Coherent method and produces a prediction of the production of type AB blood stocks per month. Based on the results against 3 years of test data, the percentage of the system accuracy rate is 100%. The reduction of learning constants and the addition of training data systems may affect the accuracy of the system in making predictions.

Time-Variant Satellite Vegetation Classification Enabled by Hybrid Metaheuristic-Based Adaptive Time-Weighted Dynamic Time Warping

Land cover data is very significant for designing the earth system, managing the natural resources, and also for performing conservation planning. Time-series data are captured with their dynamic vegetation behavior using remote sensing technology, which is broadly utilized in land cover mapping. Most of the Vegetation Index (VI) such as the Enhanced Vegetation Index (EVI) and Normalized Difference Vegetation Index (NDVI) comprises commonly employed features that are obtained from the time-series spectral data. But, these VIs are not validated as the optimal techniques for generating the temporal profiles. Recent researches highly depend on optical satellite imagery for performing these above-mentioned tasks. Dynamic Time Warping (DTW) is said to be an effective optimal solution for solving the existing challenges, especially the improved version of DTW named Time-Weighted Dynamic Time Warping (TWDTW) is used for time-series analysis regarding the time-series vegetation classification. Yet, the TWDTW efficiency is not shown with other comparative machine learning approaches owing to the classification of vegetation type in the mountain areas. The major goal of this paper is to research and create a novel approach for distinguishing the kind of vegetation in a farm region near Ujani Dam in Solapur District, Maharashtra using time-series analysis. For time-series analysis employing satellite images, the suggested model offers a unique Adaptive Time-Weighted Dynamic Time Warping (ATWDTW). The farm’s satellite images are first pre-processed before being sent to ATWDTW for examination. The TWDTW idea is optimized for classification performance using a new hybrid metaheuristic technique named Adaptive Coyote Crow Search Optimization (ACCSO). From the experimental results, the performance of the suggested ACCSO-ATWDTW correspondingly provides superior performance to the traditional approaches, where the designed model using ACCSO-ATWDTW provides 7.2%, 5.2%, 9.9%, 4.55%, and 2.33% higher MCC than the MFO-ATWDTW, BSA-ATWDTW, MF-BSA-ATWDTW, CSA-ATWDTW, and COA-ATWDTW at the maximum iteration of 200. This proved the robustness and less sensitivity to training samples of the TWDTW method when applied to mountain vegetation-type classifications.

Finite-time recurrence analysis of chaotic trajectories in Hamiltonian systems.

In this work, we show that a finite-time recurrence analysis of different chaotic trajectories in two-dimensional non-linear Hamiltonian systems provides useful prior knowledge of their dynamical behavior. By defining an ensemble of initial conditions, evolving them until a given maximum iteration time, and computing the recurrence rate of each orbit, it is possible to find particular trajectories that widely differ from the average behavior. We show that orbits with high recurrence rates are the ones that experience stickiness, being dynamically trapped in specific regions of the phase space. We analyze three different non-linear maps and present our numerical observations considering particular features in each of them. We propose the described approach as a method to visually illustrate and characterize regions in phase space with distinct dynamical behaviors.

Application of Fine-Tuned Krill Herd Algorithm in Design of Water Distribution Networks

Stochastic population-based metaheuristic algorithms have been widely used to design water distribution networks (WDNs). The present study introduces one of its kind, the krill herd algorithm (KHA), a novel swarm intelligence-based metaheuristic. Demonstrating KHA application to WDN design, the study proposes its enhanced version to improve its convergence precision. Consequently, combining the fine-tuned mechanism of KHA with the EPANET 2.2 simulation software, the fine-tuned KHA (FIT-KHA) model is formulated and checked for its efficiency in handling the WDN design problem. Because three benchmark problems (new and rehabilitation WDNs) of different dimensions are considered, the significance of the algorithm control parameters concerning the network size is studied. From sensitivity analysis, the time step factor, number of krill, and maximum iteration size are found to influence the algorithm performance. Their optimal values followed an increasing trend with the network size. The computational results reinforce a superior search exploitation ability of the FIT-KHA. Importantly, the results exhibit a better computational efficiency of KHA over most of the reported metaheuristic algorithms. Then, the performance state of optimally designed WDN under hydraulic (demand, roughness coefficient variations) and mechanical (single- and two-pipe failure) uncertain scenarios is studied. Besides reliability and surrogate measures, multiaspect metrics based on adequacy and equity are evaluated to assess the network’s performance. The WDNs are observed to be more susceptible to demand uncertainties over roughness coefficient variation. At higher states of mechanical (pipe) failure scenarios, especially, the equity of the WDN is disturbed compared to the adequacy, except for rehabilitated WDN. Certainly, the performance study manifests that the reliability measure that truly presents the WDNs performance is the function of network type. While proposing KHA as an effective alternative optimization tool, the study suggests a prior performance study to choose an appropriate metric to formulate a reliability-based multiobjective framework for the robust design of WDNs.

Optimized active contor segmentation model for medical image compression

Nowadays, medical imaging systems tend to have greatest impact on disease identification, diagnosis, and surgical preparation. To save hardware space and transmission bandwidth, it is important to reduce data redundancy in the image. Compressing images is very important for storage and transmitting purposes as it decreases the amount of bits required while retaining the critical information content encapsulated in the image document. This makes the system more peculiar in this field. The proposed paradigm image segmentation is the first step attained by the Optimized Active Contour Model (OACM). Using a new Modified marriage in honey bees optimization model (MMBO), ACM's weighting factor and maximum iteration are fine-tuned. Thereby, the collected inputimage is segmented into N-ROI and ROI. The ROI marked field will indeed be encoded using ISPIHT-based lossy compression model, whereas the non-ROI area is encoded using DCT based lossy compression model. In terms of BSC, the outcomes from both ISPIHT algorithm and DCTmodel are merged and the compressed image is itsoutput. Then, the compressed image will then be subjected to image decompression. This will include bit-stream segregation, which will be processed separately for the ROI and non-ROI regions using ISPIHT decoder and DCT-based decomposition. This process results in the original image. A comparative evaluation is undergone between the proposed and the existing techniques under PSNR, SSIM, and CR. Accordingly, the PSNR of the Proposed model is (∼)0.22, which is 26 %, 50 %, 60 %, and 55 % superior to the conventional methods like the MFO, LA, MBO, and JCF-LA.

Classification of Review Text using Hybrid Convolutional Neural Network and Gated Recurrent Unit Methods

Consumer reviews are opinions from buyers to sellers based on service satisfaction or product quality. The more consumer reviews cause the process of analyzing manually will be difficult. Therefore, an automated sentiment analysis system is needed. Each review will be grouped into a sentiment class which is divided into positive and negative classes. This study aims to classify review texts using the Convolutional Neural Network (CNN) and Gated Recurrent Unit (GRU) methods. The research stages in this study include collecting data on Tokopedia review texts, extracting hidden information from review texts using CNN, conducting learning on review texts using GRU. A total of 1000 review texts were divided into 80% training data and 20% test data. The review text is converted into matrix using One Hot Encoding algorithm and then extracted using CNN. The CNN process includes the convolution calculation, the calculation of the Rectified Linear Unit (ReLU) activation function, and the pooling stage. The extraction results in the CNN process are continued in the GRU process. The GRU process includes initializing parameters, GRU feed forward, Cross-Entropy Error calculation, GRU feedback, and updating weights and biases. The optimal weight is obtained when the error value in the training is less than the expected minimum error or the training iteration has reached the specified maximum iteration. Optimal weight is used for validation test on test data. The implementation of review text classification using the hybrid Convolutional Neural Network (CNN) and Gated Recurrent Unit (GRU) method was made using the python programming language. The accuracy of the validation test is 88.5%

An efficient algorithm and complexity result for solving the sum of general affine ratios problem

In this paper, with the unique hypothesis that the denominator is not equal to 0, an efficient outer space rectangle branch-and-bound algorithm is presented to globally solve the sum of general affine ratios problem. By using equivalent transformation and the characteristics of general single ratio function, a linear program relaxation problem is constructed for computing the lower bound of the global minimum value of the original problem. Moreover, to enhance the running speed of the presented algorithm, we design an outer space accelerating technic for deleting the entire outer space rectangle or a part of the entire examined outer space rectangle, in which there contains no the global minimum point of the equivalent problem. Furthermore, through the complexity analysis of the presented algorithm, we estimate its maximum iteration times. In addition, we prove the global convergence of the presented algorithm, report and analyze the numerical computational results for indicating the validity of the presented algorithm. Finally, two practical problems from power transportation and production planning are solved to verify the usefulness of the presented algorithm.

Improved sine algorithm for global optimization

An improved sine cosine algorithm (abbreviated MSA) is presented to overcome the disadvantages of the sine cosine algorithm (SCA), such as its low accuracy, premature convergence, and slow local convergence. We changed the method for setting the conversion parameters of the SCA algorithm from a linear decline into a nonlinear decline in order to optimize the timing of global exploration and local exploration. In order to improve the convergence accuracy and to increase the convergence speed of the SCA, an inertia weight is introduced in the position update equation. Ten high-dimensional complex benchmark functions are simulated with five improved SCA algorithms, under the same particle numbers and maximum iteration times. According to our experiments, the MSA algorithm is not only much better than two of the improved SCA algorithms in optimization accuracy but also better than the other three improved SCA algorithms in terms of stability.

Outer space branch-reduction-bound algorithm for solving generalized affine multiplicative problems

This paper presents an outer space branch-reduction-bound algorithm for globally solving generalized affine multiplicative problems (GAMP). For finding the globally optimal solution to the GAMP, we first transform the GAMP into an equivalent problem (EP) by introducing new variables and logarithmic transformation. Next, by utilizing linearization technique of logarithmic function, the affine relaxation problem of the EP can be established for obtaining the lower bound of the optimal value to the EP. Furthermore, several outer space region reduction techniques are also presented for improving the convergence speed of this algorithm. Based on outer space search and branch-and-bound framework, a branch-reduction-bound algorithm is proposed. Finally, we prove the global convergence of the proposed algorithm, and estimate maximum iterations by analyzing its computational complexity. Some experiments are reported to show the feasibility and efficiency of the algorithm.

Application of optimization algorithms for classification problem

&lt;p&gt;The work presented in this paper investigates the use of metaheuristic optimization algorithms for the face recognition problem. In the first setup, a face recognition system is implemented using particle swarm optimization (PSO) and firefly optimization algorithms, separately. PSO and firefly are used for forming the feature vectors in the feature selection stage. These feature vectors serve as the new representation for the face images that will be fed to the classifier. In the second setup, selected features from both PSO and firefly algorithms are fused to form one single feature vector for each face image before the classification stage. Extensive simulations are conducted using Poznan University of Technology (PUT) and face recognition technology (FERET) face databases. Optimal values for population size and maximum iterations number were selected before conducting the experiments. The effect of using different numbers of selected features on the performance is investigated for feature selection using PSO, firefly, and feature fusion of both.&lt;/p&gt;

A faster dynamic convergency approach for self-organizing maps

This paper proposes a novel variable learning rate to address two main challenges of the conventional Self-Organizing Maps (SOM) termed VLRSOM: high accuracy with fast convergence and low topological error. We empirically showed that the proposed method exhibits faster convergence behavior. It is also more robust in topology preservation as it maintains an optimal topology until the end of the maximum iterations. Since the learning rate adaption and the misadjustment parameter depends on the calculated error, the VLRSOM will avoid the undesired results by exploiting the error response during the weight updation. Then the learning rate is updated adaptively after the random initialization at the beginning of the training process. Experimental results show that it eliminates the tradeoff between the rate of convergence and accuracy and maintains the data's topological relationship. Extensive experiments were conducted on different types of datasets to evaluate the performance of the proposed method. First, we experimented with synthetic data and handwritten digits. For each data set, two experiments with a different number of iterations (200 and 500) were performed to test the stability of the network. The proposed method was further evaluated using four benchmark data sets. These datasets include Balance, Wisconsin Breast, Dermatology, and Ionosphere. In addition, a comprehensive comparative analysis was performed between the proposed method and three other SOM techniques: conventional SOM, parameter-less self-organizing map (PLSOM2), and RA-SOM in terms of accuracy, quantization error (QE), and topology error (TE). The results indicated the proposed approach produced superior results to the other three methods.