Proposed Hybrid Algorithm Research Articles

Nondestructive Evaluation of Thermal Barrier Coatings Thickness Using Terahertz Time-Domain Spectroscopy Combined with Hybrid Machine Learning Approaches

To ensure the thermal stability of aero-engine blades under high temperature and harsh service environments, it is necessary to quickly and accurately evaluate the thickness of thermal barrier coatings (TBCs). In this work, it was proposed to use the terahertz nondestructive testing (NDT) technique combined with the hybrid machine learning algorithm to measure the thickness of TBCs. The finite difference time-domain (FDTD) method was used to model the optical propagation characteristics of TBC samples with different thicknesses (101–300 μm) in the frequency band. To make the terahertz time-domain signal obtained simulation more realistic, uniform white noise was added to the simulation data and wavelet denoising was conducted to mimic the real testing environment. Principal components analysis (PCA) algorithm and whale optimization algorithm (WOA) combined with an optimized Elman neural network algorithm was employed to set up the hybrid machine learning model. Finally, the hybrid thickness regression prediction model shows low error, high accuracy, and an exceptional coefficient of determination R2 of 0.999. It was demonstrated that the proposed hybrid algorithm could meet the thickness evaluation requirements. Meanwhile, a novel, efficient, safe, and accurate terahertz nondestructive testing method has shown great potential in the evaluation of structural integrity of thermal barrier coatings in the near future.

Obstacle Avoidance and Path Planning of a Wheeled Mobile Robot Using Hybrid Algorithm

In the mobile robot workplace, the path planning problem is crucial. Robotic systems employ intelligence algorithms to plan the robot's path from one point to another. This paper proposes the fastest and optimal path planning of the wheeled mobile robot with collision avoidance to find the optimal route during wheeled mobile robot navigation from the start point to the target point. It is done using a modern meta-heuristic hybrid algorithm called IPSOGWO by combining Improved Particle Swarm Optimization (IPSO) with Grey Wolf Optimizer (GWO). The principal idea is based on boosting the ability to exploit in PSO with the exploration ability in GWO to the better-automated alignment between local and global search capabilities towards a targeted, optimized solution. The proposed hybrid algorithm tackles two objectives: the protection of the path and the length of the path. During, Simulation tests of the route planning by the hybrid algorithm are compared with individual results PSO, IPSO, and GWO concepts about the minimum length of the path, execution time, and the minimum number of iterations required to achieve the best route. This work's effective proposed navigation algorithm was evaluated in a MATLAB environment. The simulation results indicated that the developed algorithm reduced the average path length and the average computation time, less than PSO by (1%, 1.7%), less than GWO by (1%, 1.9%), and less than IPSO by (0.05%, 0.4%), respectively. Furthermore, the superiority of the proposed algorithm was proved through comparisons with other famous path planning algorithms with different static environments.

Hybrid optimization for optimal positioning and sizing of distributed generators in unbalanced distribution networks

Abstract The goal of this work is to reduce power loss and improve voltage profile by formulating the optimal DG placement problem as a restricted nonlinear optimisation problem. As a novelty, the proposed hybrid algorithm, referred to as Multifactor Update-based Hybrid Model (MUHM) is constructed by merging the concepts of Lion Algorithm (LA) & Sea Lion Algorithm (Sea Lion Optimization Algorithm (SLnO). The Forward-Backward Sweep (FBSM) Model is used to calculate the power loss. Three test cases are examined for the voltage profile & loss minimization in the feeder team with DGs: “case 1(DG supplying real power alone (P), case 2 (DG supplying reactive power alone (Q) and Case 3 (DG supplying both real and reactive power)”. Application of the suggested method to various IEEE test systems, including IEEE 33, IEEE 123, and IEEE 69, respectively, is used to assess its efficacy. According, the results show that the presented work at loading percentage = 0 is 12, 15, 135, 4.65, and 8 superior to SFF, BBO, BAT, LA and SLnO, respectively.

Artificial intelligence based medical decision support system for early and accurate breast cancer prediction

Feature selection, which picks the optimal subset of characteristics related to the target data by deleting unnecessary data, is one of the most important aspects of the machine learning area. A major part of big data preprocessing is feature selection (reduction). There are 2n alternative feature subsets for every n features, making it difficult to choose the best set of features from a dataset using typical feature selection techniques. Consequently, the present study proposes and suggests a unique feature selection method based on the Eagle Strategy(ESO) Optimization, Gravitational Search Optimization (GSO) algorithm, and their hybrid algorithm. We chose this infection as our subject of investigation since the number of women with breast cancer is increasing rapidly on a global scale. After lung cancer, which affects more women than any other kind of cancer, breast cancer is the second leading cause of cancer mortality. The goal of this study is to categorize breast cancer into two groups using the benchmark feature set (Wisconsin Diagnostic Breast Cancer (WDBC)) and to choose the fewest features (feature selection) to achieve maximum accuracy. This work also provides a hybrid technique for finding important features that combines two algorithms, ESO and the GSO algorithm, while reducing insignificant characteristics (features) and complexity. Soft computing technologies and machine learning algorithms provide a framework for prognostic research by classifying data instances as relevant or irrelevant depending on cancer severity. Thus, this work presented a new approach for classifying breast cancer tumors. In this research, we coupled soft computing methodologies—our implemented algorithms are applied for the first time to this problem—with artificial intelligence-based machine learning strategies to create a prediction model. The efficacy of our suggested technique was evaluated using WDBC breast cancer data sets, and the findings show that our proposed hybrid algorithm performs very well in breast cancer classification. We have been able to attain astonishing results with accuracy up to 98.9578%, sensitivity up to 0.9705, specificity up to 1.000, precision up to 1.000, F1-score up to 0.9696, and an AUC up to 0.9980 (close to maximum, i.e., 1.0000). Our study's goal is to incorporate our findings into a valid clinical prediction system, allowing visual science specialists to make more accurate and effective judgments in the future. Furthermore, our suggested technology might be used to detect a wide range of diseases.

A hybrid day-ahead electricity price forecasting framework based on time series

Electricity price forecasting (EPF) plays an indispensable role in the decision-making processes of electricity market participants. However, the complexity of electricity markets has made EPF increasingly difficult. Currently, popular methods for EPF are based on signal decomposition and suffer from computational redundancy and hyperparameter optimization challenges. In this paper, we propose a new hybrid forecasting framework to improve the forecasting accuracy of day-ahead electricity prices. The proposed model consists of three valuable strategies. First, an adaptive copula-based feature selection (ACBFS) algorithm based on the maximum correlation minimum redundancy criterion is proposed for selecting model input features. Second, a new method of signal decomposition technique for EPF field is proposed based on decomposition denoising strategy. Third, a Bayesian optimization and hyperband (BOHB) optimized long short-term memory (LSTM) model is used to improve the effect of hyperparameter settings on the prediction results. The effectiveness of the different techniques was broadly cross-validated using five datasets set up for the PJM electricity market, and the results indicated that the proposed hybrid algorithm is more effective and practical for day-ahead EPF.

Optimization of a Solar Water Pumping System in Varying Weather Conditions by a New Hybrid Method Based on Fuzzy Logic and Incremental Conductance

The present work consists of developing a new hybrid FL-INC optimization algorithm for the solar water pumping system (SWPS) through a SEPIC converter whose objective is to improve these performances. This technique is based on the combination of the fuzzy logic of artificial intelligence and the incremental conductance (INC) technique. Indeed, the introduction of fuzzy logic to the INC algorithm allows the extraction of a maximum amount of power and an improvement in the efficiency of the SWPS. The performance of the system through the SEPIC converter is compared with those of the direct coupling to show the interest of the indirect coupling, which requires an adaptation stage driven by an optimal control algorithm. In addition, a comparative analysis between the proposed hybrid algorithm and the conventional optimization techniques, namely, P&O and INC Modified (M-INC), was carried out to confirm improvements related to the SWPS in terms of efficiency, tracking speed, power quality, tracking of the maximum power point under different weather changes, and pumped water flow.

Hybrid algorithm for retinal blood vessel segmentation using different pattern recognition techniques

Blood vessel segmentation of the retina has become a necessary step in automatic disease identification and planning treatment in the field of Ophthalmology. To identify the disease properly, both thick and thin blood vessels should be distinguished clearly. Diagnosis of disease would be simple and easier only when the blood vessels are segmented accurately. Existing blood vessel segmentation methods are not supporting well to overcome the poor accuracy and low generalization problems because of the complex blood vessel structure of the retina. In this study, a hybrid algorithm is proposed using binarization, exclusively for segmenting the vessels from a retina image to enhance the exactness and specificity of segmentation of an image. The proposed algorithm extracts the advantages of pattern recognition techniques, such as Matched Filter (MF), Matched Filter with First-order Derivation of Gaussian (MF-FDOG), Multi-Scale Line Detector (MSLD) algorithms and developed as a hybrid algorithm. This algorithm is authenticated with the openly accessible dataset DRIVE. Using Python with OpenCV, the algorithm simulation results had attained an accurateness of 0.9602, a sensitivity of 0.6246, and a specificity of 0.9815 for the dataset. Simulation outcomes proved that the proposed hybrid algorithm accurately segments the blood vessels of the retina compared to the existing methodologies.

A hybrid breast cancer classification algorithm based on meta-learning and artificial neural networks.

The incidence of breast cancer in women has surpassed that of lung cancer as the world's leading new cancer case. Regular screening and measures become an effective way to prevent breast cancer and also provide a good foundation for later treatment. Women should receive regular checkups in the hospital after reaching a certain age. The use of computer-aided technology can improve the accuracy and efficiency of physicians' decision-making. Data pre-processing is required before data analysis, and 16 features are selected using a correlation-based feature selection method. In this paper, meta-learning and Artificial Neural Networks (ANN) are combined to create a hybrid algorithm. The proposed hybrid algorithm for predicting breast cancer was attempted to achieve 98.74% accuracy and 98.02% F1-score by creating a combination of various meta-learning models whose output was used as input features for creating ANN models. Therefore, the hybrid algorithm proposed in this paper can obtain better prediction results than a single model.

A cost-effective and emission-Aware hybrid system considering uncertainty: A case study in a remote area

This paper focuses on hybrid renewable energy systems (HRES) as an alternative for conventional energy systems, especially fossil fuels. The problem is to find the optimal number of components of a standalone hybrid system, based on minimizing the value of the annualized cost of the system (ACS), in the presence of the constraints. To reduce the CO2 particles as one of the most threatening particles for the planet, an additional constraint is added to the objective function. For solving the problem of system sizing, Harmony Search and Ring Theory-based Evolutionary Algorithm have drawn attention to be preferred to numerous techniques. In addition,a combination of both, which has never been used before, is proposed. The proposed hybrid algorithm shows its ability to mitigate both the emissions and costs of the system promisingly in comparison with the others. Clearly, with Carbond = 15 and different values of LPSPMax as well as using RTHS, RTEA and HS algorithms for PV/WT/BSS/DG configuration, the best performance in terms of ACS minimization belongs to RTHS algorithm with the minimum cost of $1528.2. Also, considering LPSP = 0 and changing carbon from 0 to 20 tons/year, the ACS of the PV/WT/BSS/DG configuration is reduced from $38586.3 to $21153.5 (45% reduction).

Robust, Secure and Semi-Blind Watermarking Technique Using Flexible Scaling Factor in Block-Based Wavelet Algorithm

Multimedia security has received much attention recently because of the rapid transmission of elements such as text, images, audio, video, software, animation and games. Security is becoming especially critical for content owners concerned about the illegal usage of their original products. Encryption and watermarking are two methodologies for digital applications. Spatial domain and frequency domain watermarking algorithms give very promising results in embedding binary images into the cover images. This paper proposed a new method of semi-blind watermarking technique. The digital images are divided into 4 × 4 blocks and converted into discrete Wavelet transformations (DWTs). The binary image is embedded into each block using the flexible scaling factor method. Experimental results show that the proposed method has higher peak signal to noise ratio (PSNR) and similarity ratio (SR) values compared to the standard Wavelet transformation and block-based Wavelet algorithms. The results prove that the proposed hybrid algorithm is more effective, robust, secure and resistant than DWT and block-based DWT algorithms.

Unified Evolutionary Algorithm Framework for Hybrid Power Converter

A significant amount of the literature is focused on converters that supply the required voltage with low input-current ripple to electronic devices. A hybrid converter that combines boost and Cuk converters was developed recently. This hybrid converter achieved a relatively low input-current ripple based on earlier strategies. This paper proposes a new model that simulates a hybrid power converter system using a unified evolutionary algorithm (EA). As part of this paper, we present an improved framework for a hybrid power converter. Moreover, a unified EA is developed to incorporate differential evolution (DE) and genetic algorithm (GA) properties in order to analyze the proposed modified hybrid power converter. This research describes a modified hybrid power converter that optimizes the zero-ripple duty cycle (DZ) through the proposed algorithm for minimizing the input-current ripple. Based on our simulation results, comparing the proposed method with the baseline algorithm reveals that the proposed approach is significantly more efficient than the baseline algorithm and achieves the minimum input-current ripple in different gain values. In addition, we observe that the proposed algorithm performs better than the DE and GA algorithms in terms of obtaining low input-current ripple results. Therefore, the proposed hybrid algorithm is becoming more efficient with hybridization.

A Hybrid Improved Neural Networks Algorithm Based on L2 and Dropout Regularization

Small samples are prone to overfitting in the neural network training process. This paper proposes an optimization approach based on L2 and dropout regularization called a hybrid improved neural network algorithm to overcome this issue. The proposed model was evaluated based on the Modified National Institute of Standards and Technology (MNIST, grayscale-28 × 28 × 1) and Canadian Institute for Advanced Research 10 (CIFAR10, RGB - 32 × 32 × 3) as the training data sets and data applied to the LeNet-5 and Autoencoder neural network architectures. The evaluation is conducted based on cross-validation; the result of the model prediction is used as the final measure to evaluate the quality of the model. The results show that the proposed hybrid algorithm can perform more effectively, avoid overfitting, improve the accuracy of network model prediction in classification tasks, and reduce the reconstruction error in the unsupervised domain. In addition, employing the proposed algorithm without increasing the time complexity can reduce the effect of noisy data and bias and improve the training time of neural network models. Quantitative and qualitative experimental results show that the accuracy of using the proposed algorithm in this paper with the MNIST test set has an improvement of 2.3% and 0.9% compared to L2 regularization and dropout regularization, respectively, and based on the CIFAR10 data set, the accuracy improvement of 0.92% compared with L2 regularization and 1.31% concerning dropout regularization. The reconstruction error of using the proposed algorithm in this paper with the MNIST data set has an improvement of 0.00174 and 0.00398 compared to L2 regularization and dropout regularization, respectively, and based on the CIFAR10 data set, the accuracy improvement of 0.00078 compared with L2 regularization and 0.00174 concerning dropout regularization.

Dynamic economic dispatch using hybrid CSAJAYA algorithm considering ramp rates and diverse wind profiles

• Dynamic economic dispatch performed for 10 and 15 units system. • Ramp rate and valve point effect considered for the study. • Penetration level was compared between three different wind profiles. • Novel hybrid CSAJAYA considered as optimization tool. • Results compared with those available in literature. • Statistical analysis performed for proposed hybrid algorithm. Optimal scheduling of the conventional generating units for two different dynamic test systems are percolated in this paper. This paper compares and contrasts among three types of wind profile formulations, namely linear, quadratic, and cubic, which were used to calculate wind power from hourly wind speed to find the profile with the greatest penetration of wind power. Thereafter, the wind profiles were coupled with the test system to execute dynamic economic dispatch. The optimization tool used for the study was a unique hybrid algorithm modelled by combining the properties of the recently developed crow search algorithm (CSA) and JAYA. Involvement of ramp-rate limits and the valve point effects magnified the practicality of the work thereby assessing the proposed hybrid algorithm in solving complex non-linear functions. Results infer that maximum level of wind penetration was attained by linear wind profile and a fuel cost reduction of 2.92% was realized upon incorporation of the same. Numerical results also claims that proposed hybrid CSAJAYA approach consistently yielded better quality solutions within minimum execution time without being affected by the dimension of the problem, thereby outperforming a long list of algorithms implemented for the study.

Hybrid Method for Fitting Nonlinear Height–Diameter Functions

Regression analysis is widely applied in many fields of science to estimate important variables. In general, nonlinear regression is a complex optimization problem and presents intrinsic difficulties in estimating reliable parameters. Nonlinear optimization algorithms commonly require a precise initial estimate to return reasonable estimates. In this work, we introduce a new hybrid algorithm based on the association of a genetic algorithm with the Levenberg–Marquardt method (GALM) to adjust biological nonlinear models without knowledge of initial parameter estimates. The proposed hybrid algorithm was applied to 12 nonlinear models widely used in forest sciences and 12 databases under varying conditions considering classic hypsometric relationships to evaluate the robustness of this new approach. The hybrid method involves two stages; the curve approximation process begins with a genetic algorithm with a modified local search approach. The second stage involves the application of the Levenberg–Marquardt algorithm. The final performance of the hybrid method was evaluated using total fitting for all tested models and databases, confirming the reliability of the proposed algorithm in providing stable parameter estimates. The GA was able to predict the initial parameters, which assisted the LM in converging efficiently. The developed GALM method is effective, and its application is recommended for biological nonlinear analyses.

Efficient multi-objective meta-heuristic algorithms for energy-aware non-permutation flow-shop scheduling problem

This study investigates the optimization of non-permutation flow-shop scheduling problems and lot-sizing simultaneously. Contrary to previous works, we first study the energy awareness of non-permutation flow-shop scheduling and lot-sizing using modified novel meta-heuristic algorithms. In this regard, first, a mixed-integer linear mathematical model is proposed. This model aimed to determine the size of each sub-category and determine each machine's speed within each sub-category to minimize makespan and total consumed energy simultaneously. In order to optimize this model, Multi-objective Ant Lion Optimizer (MOALO), Multi-objective Keshtel Algorithm (MOKA), and Multi-objective Keshtel and Social Engineering Optimizer (MOKSEA) are proposed. First, the validation of the mathematical model is evaluated by implementing it in a real case of the food industry using GAMS software. Next, the Taguchi design of the experiment is applied to adjust the meta-heuristic algorithms' parameters. Then the efficiency of these meta-heuristic algorithms is evaluated by comparing with Epsilon-constraint (EPC), Non-dominated Sorting Genetic Algorithm II (NSGA-II) and Multi-objective Particle Swarm Optimization (MOPSO) using several test problems. The results demonstrated that the MOALO, MOKA, and MOKSEO algorithms could find optimal solutions that can be viewed as a set of Pareto solutions, which means the used algorithm has the necessary validity. Moreover, the proposed hybrid algorithm can provide Pareto solutions in a shorter time than EPC and higher quality than NSGA-II and MOPSO. Finally, the model's key parameters were the subject of sensitivity analysis; the results showed a linear relationship between the processing time and the first and second objective functions.

A hybrid algorithm for multi-objective optimisation of parameters nanofluid in MQL-turning of Inconel X-750

ABSTRACT Inconel X-750 is referred to as hard-to-turn material in conventional machining due to the excellent mechanical and thermal properties. The present study focused on the applicability of minimum quantity of lubrication enhanced with nanoparticles to machine Inconel X-750. The nanofluid for MQL turning is 0.25 wt.% silver (Ag) + coconut oil. The process improvement of MQL is greatly dependent on MQL and machining parameters. In the L27 turning experiments, the MQL-turning parameters used were nozzle angle and distance, cutting speed, and feed rate, and the output parameters were residual stress, cutting force, and roughness. For optimisation, the Multi-Objective Golden Eagle Optimisation (MOGEO,a meta-heuristic algorithm) and Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) are utilised. Analytical hierarchy processes are utilised for weight calculation in optimisation and their effectiveness was compared. From the experimental results, the nozzle angle and distance are more significant factors for surface finish and residual stress than the force. The improvement of roughness by 6.83%, force by 47.13% and residual stress by 51.70% is observed at the optimal determined values by MOGEA rather than that of TOPSIS. The proposed hybrid algorithm of weighted MOGEO can be applied to other machining applications.

Design of Key Technologies for Elderly Public Network Services Based on Intelligent Recommendations.

As the world's population continues to increase, the proportion of elderly people is also rising. The existing elderly public service system is no longer able to meet the needs of the elderly for their daily lives. The elderly population is significantly less receptive to emerging matters than the younger population, resulting in the public elderly service system not being able to access the initial data of elderly users in a timely manner, which causes the system to make incorrect recommendations. Therefore, the elderly cannot enjoy all kinds of online services provided by the Internet platform. In order to solve this problem, an elderly intelligent recommendation method based on hybrid collaborative filtering is proposed. First, the data of elderly users and elderly service items are scored, and modelling is completed by a collaborative filtering algorithm. Then, the XGBoost model is combined to solve the optimal objective function, so that the recommended data set with the highest score in the nearest neighbour set is obtained. The experimental results show that the proposed hybrid algorithm effectively solves the cold start phenomenon that occurs when the elderly population uses the web to make recommendations for elderly services. In addition, the proposed hybrid algorithm has a higher recommendation footprint accuracy than other recommendation algorithms.

A Fully Distributed Hybrid Control Framework For Non-Differentiable Multi-Agent Optimization

This paper develops a fully distributed hybrid control framework for distributed constrained optimization problems. The individual cost functions are non-differentiable and convex. Based on hybrid dynamical systems, we present a distributed state-dependent hybrid design to improve the transient performance of distributed primal-dual first-order optimization methods. The proposed framework consists of a distributed constrained continuous-time mapping in the form of a differential inclusion and a distributed discrete-time mapping triggered by the satisfaction of local jump set. With the semistability theory of hybrid dynamical systems, the paper proves that the hybrid control algorithm converges to one optimal solution instead of oscillating among different solutions. Numerical simulations illustrate better transient performance of the proposed hybrid algorithm compared with the results of the existing continuous-time algorithms.

Modified Coral Reef Optimization Methods for Job Shop Scheduling Problems

The job shop scheduling problem (JSSP) is a fundamental operational research topic with numerous applications in the real world. Since the JSSP is an NP-hard (nondeterministic polynomial time) problem, approximation approaches are frequently used to rectify it. This study proposes a novel biologically-inspired metaheuristic method named Coral Reef Optimization in conjunction with two local search techniques, Simulated Annealing (SA) and Variable Neighborhood Search (VNS), with significant performance and finding-solutions speed enhancement. The two-hybrid algorithms’ performance is evaluated by solving JSSP of various sizes. The findings demonstrate that local search strategies significantly enhance the search efficiency of the two hybrid algorithms compared to the original algorithm. Furthermore, the comparison results with two other metaheuristic algorithms that also use the local search feature and five state-of-the-art algorithms found in the literature reveal the superior search capability of the two proposed hybrid algorithms.

Automatic Data Clustering by Hybrid Enhanced Firefly and Particle Swarm Optimization Algorithms

Data clustering is a process of arranging similar data in different groups based on certain characteristics and properties, and each group is considered as a cluster. In the last decades, several nature-inspired optimization algorithms proved to be efficient for several computing problems. Firefly algorithm is one of the nature-inspired metaheuristic optimization algorithms regarded as an optimization tool for many optimization issues in many different areas such as clustering. To overcome the issues of velocity, the firefly algorithm can be integrated with the popular particle swarm optimization algorithm. In this paper, two modified firefly algorithms, namely the crazy firefly algorithm and variable step size firefly algorithm, are hybridized individually with a standard particle swarm optimization algorithm and applied in the domain of clustering. The results obtained by the two planned hybrid algorithms have been compared with the existing hybridized firefly particle swarm optimization algorithm utilizing ten UCI Machine Learning Repository datasets and eight Shape sets for performance evaluation. In addition to this, two clustering validity measures, Compact-separated and David–Bouldin, have been used for analyzing the efficiency of these algorithms. The experimental results show that the two proposed hybrid algorithms outperform the existing hybrid firefly particle swarm optimization algorithm.