Random Optimization Algorithm Research Articles

Structural optimization design of multi ribbed composite wall of building components under seismic load based on random optimization algorithm and resilience model

The multi ribbed composite wall structure is also known as the multi ribbed wall panel light frame structure. This structure is suitable for housing construction in the residential field. The special structural failure process and mode of multi ribbed composite walls are different from traditional walls. To fully utilize the excellent structural performance in building construction and improve the seismic performance of the building, based on the transformation principle of subset optimization algorithm for optimization problems, a constrained subset simulation optimization algorithm suitable for optimizing the maximum displacement angle of multi ribbed composite wall panels is designed. The Bayesian algorithm is used to construct a restoring force model for multi ribbed composite wall panels. The constrained subset simulation optimization algorithm and resilience model are used to optimize the seismic performance of 4-layer multi ribbed composite wall panels. The results show that the section height and the equivalent slant support width of the continuous column for the 4-story multi ribbed composite wall panel change from discrete distribution to aggregation with the increase of iteration. Finally, the sampling is stable in the 9th floor. At this time, the section height of the continuous column is 230 mm, and the equivalent slant support width is 525. After optimization, the failure probability of both extreme displacement angle states has decreased. When the peak ground acceleration is 1.0 g, the optimized second limit state failure probability is less than 100%. When the peak ground acceleration value is between 0.2 g and 0.6 g, both limit states show a rapid upward trend. The constrained subset simulation optimization algorithm and Bayesian quantitative resilience model proposed in the research can effectively optimize the seismic performance of multi ribbed composite walls.

Ensemble Random Forest-based Gradient Optimization based Energy Efficient Video Processing System for Smart Traffic Surveillance System

Deep learning solutions in big data applications can benefit cloud centres and can also lead to network communication overhead. Typically, data collected from traffic are sent to the traffic management centre for analysis. However, this process can worsen the network route to the traffic management centre. A two-tier mechanism has been developed to address this issue, which performs vehicle speed estimation and traffic congestion detection for efficient traffic management. The real-time traffic video data are captured and the video frames are initially processed through a foreground extraction process, which extracts the temporarily stopped vehicles on the road by removing background pixels from the frames. The video frames are then wrapped in an up-down view to remove the influence of the observation angle. The traffic congestion is then detected accurately based on the traffic characteristics using the proposed Ensemble Random Forest-based Gradient Optimization (ERF-GO) algorithm. The generalization error occurs when learning complex features on frames is minimized using a gradient-based optimization (GO) algorithm. Finally, the learned information on traffic conditions is forwarded to the cloud and edge computing environments based on network connection speed. The efficiency of the proposed ERF-GO is investigated in terms of performance metrics, namely root mean square error, speed detection error, execution time, computational cost, accuracy, latency, workload balance, precision, recall, f-measure, and congestion detection error rate. The analytic result displays that the proposed ERF-GO algorithm attains a greater accuracy rate of about 98.65% in detecting traffic congestion which is comparably higher than state-of-the-art methods.

Fundus image segmentation based on random collision whale optimization algorithm

Medical image segmentation is an important technical tool, OTSU algorithm is a common method in threshold segmentation, but with the increase of the number of threshold segmentation, the selection of its threshold is a big problem, and the segmentation effect is difficult to be guaranteed. In order to solve this problem, this paper proposes a random collision whale optimization algorithm to optimize OTSU for reliable image segmentation. The algorithm is called RCWOA for short. Firstly, the Halton sequence is used to uniformly initialize the population to make the population position distribution uniform, and then the dimensional Opposition-based learning of small-hole imaging is introduced to update the whale position and find out the missing feasible solution. Finally, the random collision theory is used to update the position of the optimal individual to improve the quality of the solution, At the same time, it also improves the search ability of the algorithm. In 12 test functions, RCWOA was compared with 6 other algorithms, demonstrating the feasibility and novelty of RCWOA. In 8 experiments of fundus image segmentation, RCWOA was compared with 9 other algorithms. The results showed that RCWOA had a Friedman test composite ranking of 1.3516, ranking at the forefront, and exhibited significantly improved segmentation quality.

VoteDroid: a new ensemble voting classifier for malware detection based on fine-tuned deep learning models

AbstractIn this work, VoteDroid a novel fine-tuned deep learning models-based ensemble voting classifier has been proposed for detecting malicious behavior in Android applications. To this end, we proposed adopting the random search optimization algorithm for deciding the structure of the models used as voter classifiers in the ensemble classifier. We specified the potential components that can be used in each model and left the random search algorithm taking a decision about the structure of the model including the number of each component that should be used and its location in the structure. This optimization method has been used to build three different deep learning models namely CNN-ANN, pure CNN, and pure ANN. After selecting the best structure for each DL model, the selected three models have been trained and tested using the constructed image dataset. Afterward, we suggested hybridizing the fine-tuned three deep-learning models to form one ensemble voting classifier with two different working modes namely MMR (Malware Minority Rule) and LMR (Label Majority Rule). To our knowledge, this is the first time that an ensemble classifier has been fine-tuned and hybridized in this way for malware detection. The results showed that the proposed models were promising, where the classification accuracy exceeded 97% in all experiments.

Fast random opposition-based learning Aquila optimization algorithm

Meta-heuristic algorithms are usually employed to address a variety of challenging optimization problems. In recent years, there has been a continuous effort to develop new and efficient meta-heuristic algorithms. The Aquila Optimization (AO) algorithm is a newly established swarm-based method that mimics the hunting strategy of Aquila birds in nature. However, in complex optimization problems, the AO has shown a sluggish convergence rate and gets stuck in the local optimal region throughout the optimization process. To overcome this problem, in this study, a new mechanism named Fast Random Opposition-Based Learning (FROBL) is combined with the AO algorithm to improve the optimization process. The proposed approach is called the FROBLAO algorithm. To validate the performance of the FROBLAO algorithm, the CEC 2005, CEC 2019, and CEC 2020 test functions, along with six real-life engineering optimization problems, are tested. Moreover, statistical analyses such as the Wilcoxon rank-sum test, the t-test, and the Friedman test are performed to analyze the significant difference between the proposed algorithm FROBLAO and other algorithms. The results demonstrate that FROBLAO achieved outstanding performance and effectiveness in solving an extensive variety of optimization problems.

An effective framework of human abnormal behaviour recognition and tracking using multiscale dilated assisted residual attention network

The prevention of certain unwanted crime events and eliminating them even before their execution can be done by automatic identification of abnormal behavior in humans. Hence automatic prediction of abnormal human behavior is a difficult task to perform. Some of the automated model has been implemented and provided the most promising results. The manual intervention is being the greatest approach in earlier time, yet it brings with numerous errors, consumes more time and more cost effective. Henceforth, the automated model is suggested for identifying the activities. As the scholar focus on machine and deep learning, this classifier may extract the hand-crafted features. But it fails to yield the appropriate solution for finding the activities. Since it belongs to the video frames, the object detection is highly Ineffective feature vector and inadequate scale measures of the learning model paves the way for performance degradation. This issue can be resolved by including an attention mechanism in the deep learning model for both monitoring and classification purposes. The recommended Human Abnormal Behavior Recognition and Tracking (HABRT) model performs the following operations, such as the collection of video, categorizing the behavior in the video as normal or abnormal, monitoring, extraction of the object, and classification of the abnormality. The input video with such frames is initially gathered from publically available databases. By using these frames, the abnormal behavior classification is done by Multiscale Dilated assisted Residual Attention Network (MD-RAN), For further enhancement, the hyper-parameters in the MD-RAN are optimally selected by novel Modified Random Parameter-based Chimp Optimization Algorithm (MRP-ChOA). Once the abnormal frames are obtained, the activity tracking is achieved by Adaptively Modified You Only Look Once (YOLO) V3 (AM-YOLO V3). This model encompasses with multiple layers, so that utilized number of layers are determined optimally using MRP-ChOA. Consequently, the objects are extracted from the abnormal frames with the help of AM-YOLO V3. Finally, the abnormalities are classified by using the same MD-RAN. At last, the performance is analyzed and validated with diverse parameters, which are then compared with other algorithms. While implementing the dataset 1, the accuracy value attains maximum in contrast with 3.14% of DTCN, 2.308% of CNN-RNN and 13.7% of ResAttenConvLSTM, correspondingly. Thus, the findings reveal that it has the potential to deliver extensive results for abnormal recognition and tracking.

Research on kinematics modeling and path planning of a hyper-redundant continuum robot

It has always been the case that work in confined spaces has high difficulty and complexity, and robots are usually required to replace humans to perform the corresponding operations. For the maintenance work in the fuel tank of a certain type of aircraft, a continuum robot with 32 staggered orthogonal joints is designed in this paper. To solve the motion problem of this robot, this paper combines the end-following method with the Jacobi inverse kinematics method, and proposes an end-following-segmented Jacobi method, which improves the accuracy and efficiency of the inverse kinematics solution. A ray path-planning algorithm for continuum robots is proposed for the job requirements of this robot and the specific operating environment inside an airplane fuel tank, and it is compared with the rapidly-exploring random trees (RRT) algorithm and particle swarm optimization (PSO) algorithm. The experimental results show that the planning time of the ray path-planning algorithm is less compared with the RRT algorithm and the PSO algorithm, and the paths obtained are shorter. Finally, a path-following simulation is performed on a 32-joint continuum robot model to verify the followability of the ray path algorithm.

A Novel Heart Disease Prediction System using Deep Multi-Layer Perceptron and Optimal Feature Selection Mechanism

Diagnosis and prognosis of heart disease (HD) are essential medical tasks for a correct classification, which helps cardiologists to treat the patient properly. The current medical system is unable to obtain the entire information from the heart disease database. It is difficult for a physician to analyze and diagnose chronic disease because it is a challenging endeavor. Hence this paper proposes a novel weight and bias tune deep multi-layer perceptron for heart disease prediction (WBTDMLP) with optimal feature selection using modified random forest (MRF). The proposed system comprised ‘3’ phases such as data preprocessing, feature selection, and HD prediction. Initially the HD prediction data is collected from the Cleveland dataset and the missing value imputation and data normalization is applied on the dataset to preprocess the dataset. Following that, the feature selection was performed by using the MRF algorithm. Finally, the HD prediction is done based on WBTDMLP approach and the parameters are tuned by Sobel sequence with Brownian random walk-based dragonfly optimization algorithm (SSBRWDOA). The results indicate that the proposed approach reaches 97.89% accuracy, which is relatively higher than existing methods.

Software reliability prediction using ensemble learning with random hyperparameter optimization

The paper investigates software reliability prediction by using ensemble learning with random hyperparameter optimization. Software reliability is a significant problem with software quality that developers face. It involves accurately predicting the next failure. In recent years, machine learning techniques and ensemble learning approaches have been applied to improve software reliability prediction. These approaches aim to analyze historical data and develop models that can accurately forecast when failures are likely to occur. The article proposes an ensemble learning regression model using Ridge, Bayesian Ridge, Support Vector Regressor (SVR), K-Nearest Neighbors Algorithm (KNN), Regression tree, Random Forest, Neural network, and Decision Tree as base learners. Ridge is used as a combiner model. Each base learner hyperparameter is tuned using a random search algorithm automatically. A random hyperparameter search optimization algorithm selects the hyperparameter and adjusts it for overfitting and underfitting. The base models are tuned to minimize bias and variance. The performances of the models are evaluated using standard error measures such as Mean Squared Error (MSE), Sum of Squared Error (SSE), and Normalized Root Mean Square Error (NRMSE). The proposed ensemble model is compared with existing models using a benchmark dataset. The Iyer,and Lee, and Musa datasets are used for the experiment. The dataset is scaled using standard methods like logarithmic scaling, lagging, and linear interpolation. The results of the statistical comparison show better performance by our proposed model as compared to existing models.

Path Planning Using Combined Informed Rapidly- Exploring Random Tree Star and Particle Swarm Optimization Algorithms

Path Planning Using Combined Informed Rapidly- Exploring Random Tree Star and Particle Swarm Optimization Algorithms

Total harmonic distortion optimization in a seven level multilevel inverter by the random search heuristic algorithm

In this article, the random search heuristic optimization algorithm is proposed to guarantee that the voltage synthesized by a seven-level multilevel inverter contains the lowest possible total harmonic distortion. The algorithm in this work is presented as an alternative, simple, and easy-to-implement method for solving a set of transcendental mathematical equations of multilevel inverters. The obtained results demonstrate the algorithm's capability to solve the mathematical formulation that minimizes the total harmonic distortion. It is important to highlight that the algorithm was implemented in Matlab®, and the obtained results were validated in a simulation conducted in Simulink®. Equally and not less important, the results were physically implemented in a laboratory prototype of three integrated circuits with H-bridges containing insulated gate bipolar transistors.

Recognition of Arabic Air-Written Letters: Machine Learning, Convolutional Neural Networks, and Optical Character Recognition (OCR) Techniques

Air writing is one of the essential fields that the world is turning to, which can benefit from the world of the metaverse, as well as the ease of communication between humans and machines. The research literature on air writing and its applications shows significant work in English and Chinese, while little research is conducted in other languages, such as Arabic. To fill this gap, we propose a hybrid model that combines feature extraction with deep learning models and then uses machine learning (ML) and optical character recognition (OCR) methods and applies grid and random search optimization algorithms to obtain the best model parameters and outcomes. Several machine learning methods (e.g., neural networks (NNs), random forest (RF), K-nearest neighbours (KNN), and support vector machine (SVM)) are applied to deep features extracted from deep convolutional neural networks (CNNs), such as VGG16, VGG19, and SqueezeNet. Our study uses the AHAWP dataset, which consists of diverse writing styles and hand sign variations, to train and evaluate the models. Prepossessing schemes are applied to improve data quality by reducing bias. Furthermore, OCR character (OCR) methods are integrated into our model to isolate individual letters from continuous air-written gestures and improve recognition results. The results of this study showed that the proposed model achieved the best accuracy of 88.8% using NN with VGG16.

Temperature prediction of Bragg grating sensing based on a one-dimensional convolutional neural network.

In this paper, we propose a new (to us) way of demodulating the grating sensing spectrum using a one-dimensional convolutional neural network (1DCNN) to overcome the limitation of the traditional fitting method of temperature demodulation for subway tunnel fires. This method constructs a one-dimensional convolutional neural network model and combines it with the experimental device of a fiber Bragg grating (FBG) temperature measurement. One thousand eight hundred spectra of experimental data are selected as sample data for training. Adam's random optimization algorithm is used in training to predict the temperature of multiple periods, with an accuracy of 99.95% and a root-mean-square deviation (RMSE) of 0.0832°C. The experiment shows that the algorithm in this paper is better than the GRU and LSTM algorithms, as traditional maximum peak methods, and can effectively improve the measurement accuracy. This article aims to provide a high-speed demodulation solution for FBG-based sensing systems to meet the practical needs of large-scale real-time monitoring.

Efficient generation of random fiber distribution by combining random sequential expansion and particle swarm optimization algorithms

Representative volume elements (RVEs) with random fiber distribution are widely used in micro-mechanics for determining the properties of unidirectional fiber-reinforced composites from their microstructure. In this paper, the random sequential expansion (RSE) and particle swarm optimization (PSO) algorithms are combined to develop an efficient methodology for generating such RVEs. This methodology can successfully eliminate the biased regions of dense fiber population and unreal matrix-rich corners that usually appear in created RVEs and resolve the complications in selection of the input parameters in the RSE algorithm. Statistical analysis of the generated microstructures and two- and three-dimensional finite element analyses are performed to validate the capability of the proposed algorithms.

Biomarker identification and cancer survival prediction using random spatial local best cat swarm and Bayesian optimized DNN

Identifying cancer biomarkers is crucial for improving patient outcomes and reducing cancer-related deaths. This research proposes BioSurv, a framework for biomarker identification and cancer survival prediction, using machine learning and deep learning techniques. Multi-omics data from breast cancer (BRCA) and lung adenocarcinoma (LUAD), including mRNA, miRNA, CNV, and DNA methylation, are analyzed. The collected dataset is passed to statistical tests and the random spatial local best cat swarm optimization (RSLBCSO) algorithm for feature selection, followed by KEGG and survival analyses to identify prognostic markers. Thirteen BRCA and fifteen LUAD poor prognostic markers are identified. A Bayesian optimized deep neural network (DNN) is used for cancer survival prediction, achieving high accuracy of 90% and 91% for BRCA and LUAD, respectively.

NPROS: A Not So Pure Random Orthogonal search algorithm—A suite of random optimization algorithms driven by reinforcement learning

NPROS: A Not So Pure Random Orthogonal search algorithm—A suite of random optimization algorithms driven by reinforcement learning

Study of Pricing of High-Dimensional Financial Derivatives Based on Deep Learning

Many problems in the fields of finance and actuarial science can be transformed into the problem of solving backward stochastic differential equations (BSDE) and partial differential equations (PDEs) with jumps, which are often difficult to solve in high-dimensional cases. To solve this problem, this paper applies the deep learning algorithm to solve a class of high-dimensional nonlinear partial differential equations with jump terms and their corresponding backward stochastic differential equations (BSDEs) with jump terms. Using the nonlinear Feynman-Kac formula, the problem of solving this kind of PDE is transformed into the problem of solving the corresponding backward stochastic differential equations with jump terms, and the numerical solution problem is turned into a stochastic control problem. At the same time, the gradient and jump process of the unknown solution are separately regarded as the strategy function, and they are approximated, respectively, by using two multilayer neural networks as function approximators. Thus, the deep learning-based method is used to overcome the “curse of dimensionality” caused by high-dimensional PDE with jump, and the numerical solution is obtained. In addition, this paper proposes a new optimization algorithm based on the existing neural network random optimization algorithm, and compares the results with the traditional optimization algorithm, and achieves good results. Finally, the proposed method is applied to three practical high-dimensional problems: Hamilton-Jacobi-Bellman equation, bond pricing under the jump Vasicek model and option pricing under the jump diffusion model. The proposed numerical method has obtained satisfactory accuracy and efficiency. The method has important application value and practical significance in investment decision-making, option pricing, insurance and other fields.

Fast random opposition-based learning Golden Jackal Optimization algorithm

Nowadays, optimization techniques are required in various engineering domains in order to find optimal solutions for complex problems. As a result, there is a growing tendency among scientists to enhance existing nature-inspired algorithms using various evolutionary strategies and to develop new nature-inspired optimization methods that can properly explore the feature space. The recently designed nature-inspired metaheuristic, named the Golden Jackal Optimization​ (GJO) algorithm, was inspired by the collaborative hunting actions of the golden jackal in nature to solve various challenging problems. However, like other approaches, the GJO has the limitations of poor exploitation ability, ease to get stuck in a local optimal region, and an improper balancing of exploration and exploitation. To overcome these limitations, this paper proposes a novel contribution to GJO based on a new technique, namely the fast random opposition-based learning Golden Jackal Optimization algorithm (FROBL-GJO). The FROBL technique is mainly inspired by opposition-based learning (OBL) and random opposition-based learning (ROBL) techniques to enhance the optimization precision and convergence speed of the GJO algorithm. Furthermore, two other models, such as OBL-GJO and ROBL-GJO, are also proposed for comparison purposes. To examine the proficiency of the newly proposed FROBL-GJO algorithm, it has been examined with several well-known existing meta-heuristic algorithms while solving the CEC-2005 and CEC-2019 benchmark test functions and real-life engineering problems. The experimental outcomes and statistical tests reveal the superior performance of the proposed FROBL-GJO in solving both global optimization and engineering design problems. Hence, the findings of benchmark functions and engineering problems endorse that the proposed FROBL-GJO algorithm can be considered a promising method for solving complex optimization problems.

A Novel Image Recovery from Moving Water Surface Using Multi-Objective Bispectrum Method

Nowadays, the image degradation field suffers from several challenges while processing underwater color images including color distortion and image blurring due to the scattering media. Moreover, to get appropriate multi-frame super-resolution images, there is essential for recovering a better quantity of images. Traditionally, the shift among images is directly evaluated when considering the under-sampled Low-Resolution (LR) images. On the other hand, the high-frequency LR image faces unreliability owing to the aliasing consequences of sub-sampling, but it will also degrade the recovery accuracy. This task design implements a novel image recovery model from the moving water surface by adopting the multi-objective adaptive higher-order spectral analysis. Image pre-processing, lucky region selection, and image recovery are the three main phases of this model. The bicoherence method and dice coefficient method are adopted for performing the lucky region selection. Finally, the adoption of the multi-objective adaptive bispectra method is used for performing the image recovery from the moving water surface. The improved Adaptive Fitness-oriented Random number-based Galactic Swarm Optimization (AFR-GSO) algorithm is used for optimizing the constraints of the bispectrum method. The experimental results verify the enrichment of image quality by the proposed model over the existing techniques.

Multi-objective optimal peak load shaving strategy using coordinated scheduling of EVs and BESS with adoption of MORBHPSO

With increase of electrical vehicles (EVs), carbon emissions and dependence on fossil energy would be reduced. However, uncoordinated charging may further raise the peak load and magnify the load imbalance which will pose a key challenge to the system reliability. Hence, the peak load shaving whereby coordinated optimal scheduling of EVs and energy storage systems (ESS) has attracted more and more attention. And challenges arise in terms of multiple objectives, algorithms with better global searching performance and constraints handling methods. In this paper, the proposed multi-objective optimal peak load shaving strategy aims to achieve the best peak load shaving effect with the minimum electricity cost whereby coordinated scheduling of EVs and battery energy storage systems (BESS). Especially, load balance constraint, charging/discharging power limits considering state of charge (SOC) as well as capacity limits of EVs and BESS, vehicles to grid (V2G), time-of-use (TOU) price and driving behavior of EVs with different types are all considered. The multi-objective random black-hole particle swarm optimization algorithm (MORBHPSO) with adjustable power redundancy method is adopted. Four case studies have been carried out on a regional distribution network with 130 EVs and 20 BESS. And satisfactory results were obtained in terms of better peak load shaving effect (a 70.6 % decrease in load fluctuation level) and better economic benefit (a 40.56 % reduction in electricity cost). Moreover, MORBHPSO performs better than multi-objective particle swarm optimization algorithm (MOPSO) in terms of a 41.47 % decline in load fluctuation level and a 5.44 % decrease in electricity cost with only half the iterations. Furthermore, it is also found that different from the impact of EVs' driving behavior, TOU price is conducive to obtaining smoother load curve and lower electricity cost.