Technique Particle Swarm Research Articles

Comparism of Response Surface Methodology (RSM) and Adaptive Neuro-Fuzzy Inference Systems (ANFIS) in Optimisation of Soybean Soapstock Biodiesel Production

Soybean soapstock (SS), a lipid rich by-product of soybean oil production is a promising feedstock for the production ofbiodiesel due to its availability and affordability. In the esterification and transesterification reactions involving soyabeansoapstock, sodium hydroxide, methanol and n-hexane were used as catalyst, solvent and co-solvent respectively. The physico-chemical properties of the biodiesel obtained were determinedusing the Association of Analytical Chemist (AOAC) and American Society of Testing Materials (ASTM) methods. The esterification and transesterification reactions were optimised using both response surface methodology (RSM) under design expert 7.0 platform and Particle swarm technique in ANFIS (ANFIS-PSO) using the MATLAB software. The optimized acid value from the esterification reaction using RSM and ANFIS-PSO were 4.956 and 1.488 while the yield obtained were 97.29% and 99.91%respectively with ANFIS-PSO proving to be the better optimization technique in both cases. Comparison plots made for both reactions shows the ANFIS-PSO curve mirroring the experimental and thus signifying a closer trend when compared to the RSM curve. The suitability of the ANFIS-PSO prediction was further highlighted by the error analysis carried out on both techniques. The Residual sum of squares (RSS), Mean absolute error (MAE), Root mean square error (RMSE), Correlation coefficient (R), Coefficient of determination (R<sup>2</sup>), Adjusted R<sup>2</sup>, Absolute average deviation (AAD) and Mean absolute percent error (MAPE) values for the ANFIS-PSO predictions in both reactions were better than the RSM predictions. It can thus be concluded that soybean soapstock is a viable feedstock for biodiesel production and ANFIS-PSO is a more efficient optimization technique when compared with RSM in esterification and transesterification of soybean soapstock.

Optimizing Solar-Wind Hybrid Microgrid Designs with Particle Swarm Techniques for Sustainable Energy Integration

Optimizing Solar-Wind Hybrid Microgrid Designs with Particle Swarm Techniques for Sustainable Energy Integration

Ultrasound-assisted extraction of phytochemicals from green coconut shell: Optimization by integrated artificial neural network and particle swarm technique

This study employs artificial neural network (ANN) and particle swarm optimization (PSO) to maximize antioxidant and antimicrobial activity from green coconut shells. Phytochemical analysis was carried out on the extract obtained from ultrasound-assisted extraction performed at different combinations of time (10, 20, and 30 min), temperature (30, 35, and 40 °C), and the ratio of solid-solvent (1:10, 1:20, and 1:30 g/ml). The presence of these bioactive compounds exhibits antimicrobial and antioxidant activities. Quantitative analysis showed that the total phenolic compounds ranged from 7.08 to 33.46 mg GAE/g, flavonoids ranged from 2.09 to 28.46 mg QE/g, tannins ranged from 70.5 to 141.09 mg TAE/g, and antioxidant activity of 49.98–66.1 %. The FTIR analysis detected the presence of CO, O–H, and C–H bonds. The optimized condition of ultrasound-assisted extraction (UAE) was compared with the optimized condition of the microwave. The result of ultrasound-assisted extraction was observed to be better than microwave-assisted extraction.

Optimization of a multistage liquid desiccant dehumidifier by neural networks & particle swarm techniques

To overcome the inconsistent and limited dehumidification by single stage dehumidification system, a novel multistage, reciprocating dynamic liquid desiccant dehumidification system has been proposed to dehumidify larger space consistently. Experimental investigation on the proposed unit has been conducted by varying air velocity, cam shaft speed, and inlet relative humidity (RH). System's performance in terms of moisture removal rate (MRR), dehumidification efficiency (DE) and energy consumed has been compared with that of a single stage system. It is observed that multistage system performed better than the single stage system in terms higher MRR and efficiency. To optimise the system performance, neural network-based Particle Swarm optimization (PSO) has been employed. Inertia weight distribution for PSO has been obtained by the statistical Taguchi method. PSO optimized results for randomly distributed weights have been compared with the Taguchi given weights. It is found that the performance of the PSO with this method is better than the former in terms of optimized paramters. PSO predicted optimum input variables are found as air velocity of 8.2 m/s, inlet RH of 90% and cam speed of 15.16 rpm. The maximum performance parameters obtained are MRR of 7.549 g/s, Dehumidification efficiency of 0.762 and energy consumption of 2165.4 Watts. Confirmation experiments indicated that PSO predicted results are in well agreement with the experimental results with a maximum deviation of 1.8%. Current work predicts and optimizes the dehumidification performance which minimizes experimentation time and also the cost involved.

Numerical approach based on particle swarm technique for optimization of nonlinear buried pipeline on tensionless foundation

This work aims to contribute to the development of numerical approach for optimization of buried pipeline by using finite element (FE) method and particle swarm optimization (PSO), considering static analysis and material hardening. The buried pipeline is discretized by finite element approach, constituted by Euler-Bernoulli three nodes beam element, while the soil-structure interaction is modeled by Winker-type tensionless foundation. Moreover, the tensionless foundation is discretized by spring element in the numerical model. The material nonlinearity behavior is analyzed by Von Mises isotropic hardening model. In the numerical approach, this model is solved by stress increment procedure. Furthermore, the critical loading and optimum wall thickness, that leads to optimized condition, are determined using swarm techniques. The cost function is formulated using evaluated stress and target stress, more, the calculated error is minimized by using swarm techniques for single-variable and two-variable optimization. Several applications are carried out by considering several soil conditions, boundary conditions and loading types. As a novelty, the soil random stiffness is included in the numerical model, where the optimum wall thickness and critical loading are evaluated considering material nonlinearity. The performance of finite element particle swarm technique is also compared to finite element genetic algorithms technique. The results demonstrate the effectiveness of proposed numerical approach for nonlinear buried pipeline optimization.

Application Analysis of Intelligent Particle Swarm Algorithm in the Development of Modern Tourism Intelligence

Information technology has been increasingly vital in the rapid expansion of the tourism industry economy. The new information technology has become a significant driving force and factor in the tourism industry’s economic development. The application of new information technology to the advancement of modern tourist intelligence has emerged as an essential study area. Smart tourism, aided by informatization, can frequently provide more attractive economic benefits to tourism businesses. Tourism is being positioned as a new economic growth point and core business in an increasing number of cities. Various measures have been taken among the main body of the city to improve the tourism competitiveness of the city, so as to be in an advantageous position in the fierce market competition. Under this background, the research on urban tourism competitiveness has become a research hotspot in the development of modern tourism intelligence. This study used a combined PSO and neural network to assess the competitiveness of urban tourism. First, this paper offers a method that makes use of an improved IPSO. It changes the inertia weight dynamically and nonlinearly based on the particle fitness value. Simultaneously, the particle swarm technique is enhanced by combining the particle iterative cycle to boost position disturbance. Second, in order to build an IPSO-BP network, this work optimizes the initial weights and thresholds of the BP network based on IPSO. This work uses this network to evaluate the competitiveness of urban tourism, which can overcome the defects of traditional BP network. Third, this work conducts systematic experiments on the IPSO-BP method, and the experimental results confirm the superiority of this method in the evaluation of urban tourism competitiveness.

Improving Images in Turbid Water through Enhanced Color Correction and Particle Swarm-Intelligence Fusion (CCPF)

When light travels through a water medium, selective attenuation and scattering have a profound impact on the underwater image. These limitations reduce image quality and impede the ability to perform visual tasks. The suggested integrated color correction with intelligence fusion of particle swarm technique (CCPF) is designed with four phases. The first phase presents a novel way to make improvement on underwater color cast. Limit the improvement to only red color channel. In the second phase, an image is then neutralized from the original image by brightness reconstruction technique resulting in enhancing the image contrast. Next, the mean adjustment based on particle swarm intelligence is implemented to improve the image detail. With the swarm intelligence method, the mean values of inferior color channels are shifted to be close to the mean value of a good color channel. Lastly, a fusion between the brightness reconstructed histogram and modified mean particle swarm intelligence histogram is applied to balance the image color. Analysis of underwater images taken in different depths shows that the proposed CCPF method improves the quality of the output image in terms of neutralizing effectiveness and details accuracy, consequently, significantly outperforming the other state-of-the-art methods. The proposed CCPF approach produces highest average entropy value of 7.823 and average UIQM value of 6.287.

A hybrid swarm intelligence feature selection approach based on time-varying transition parameter

Feature selection aims to reduce the dimensionality of a dataset by removing superfluous attributes. This paper proposes a hybrid approach for feature selection problem by combining particle swarm optimization (PSO), grey wolf optimization (GWO), and tournament selection (TS) mechanism. Particle swarm enhances the diversification at the beginning of the search mechanism, grey wolf enhances the intensification at the end of the search mechanism, while tournament selection maintains diversification not only at the beginning but also at the end of the search process to achieve local optima avoidance. A time-varying transition parameter and a random variable are used to select either particle swarm, grey wolf, or tournament selection techniques during search process. This paper proposes different variants of this approach based on S-shaped and V-shaped transfer functions (TFs) to convert continuous solutions to binaries. These variants are named hybrid tournament grey wolf particle swarm (HTGWPS), followed by S or V letter to indicate the TF type, and followed by the TF’s number. These variants were evaluated using nine high-dimensional datasets. The results revealed that HTGWPS-V1 outperformed other V’s variants, PSO, and GWO on 78% of the datasets based on maximum classification accuracy obtained by a minimal feature subset. Also, HTGWPS-V1 outperformed six well-known-metaheuristics on 67% of the datasets.

Multi-objective optimization of a hybrid system based on combined heat and compressed air energy storage and electrical boiler for wind power penetration and heat-power decoupling purposes

Faced with uncertain heating supply task in heating season, the combined heat and power (CHP) unit groups in “Three North Areas” of China with considerable capacity of wind power often operate in heat-controlled mode with limited flexibility, which exacerbates the wind power curtailment. In order to further improve the operation flexibility of CHP unit group and the wind power penetration level, a hybrid system based on CHP unit group embedded with electrical boiler subsystem and combined heat and compressed air energy storage (CH-CAES) subsystem is proposed. A novel operation strategy on power generation side considering device reserve constraints and transient behaviors is put forward. Then, a bi-level optimization algorithm based on multi-objective particle swarm optimization (MOPSO) and technique for order preference by similarity to an ideal solution (TOPSIS) method is utilized. Finally, a completed heat season case is conducted to investigate the dynamic behaviors and the performance of the improved system. The simulation results of optimal plan mainly dominated by ratio of abandoned wind (ROAW) show that energy and exergy efficiencies with remaining stored energy can be improved by 3.52 % and 2.86 % compared with the basic system, respectively. For the economic performance, the levelized cost of heat (LCOH) drops by 43.56 % while levelized cost of electricity (LCOE) increases by 4.99 %. In the environmental aspect, the coal consumption, CO2 and SO2 emission can reduce by 7.35 %, 7.34 % and 7.35 %, respectively. The result shows that the improved system can decrease the curtailed wind power significantly and boost the CHP operation flexibility remarkably. Meanwhile, the dynamic behaviors of system components in reasonable scope mean that the fluctuant renewable energy can be eliminated efficiently, validating the effectiveness of the proposed system in high shared wind region simultaneously.

An improved pathfinder algorithm using opposition-based learning for tasks scheduling in cloud environment

Cloud computing (CC) has become a major topic of study in recent years, providing a cost-effective deployment framework for hosting and running workflows. It is considered as the leading model for distributed computing due to its elasticity, speed, and pay-per-use model. Running scientific applications in the cloud environment with a number of criteria remains a great challenge for researchers, because it requires a robust infrastructure with high computing performance, communication and storage capabilities. Most previous works designed for cloud environments examine the fixed number of resources and aims only to reduce runtime. This has forced scientists to develop the workflow scheduling (WFS) optimization algorithms that strike the right balance between the three main qualities of service (QoS) parameters: time, cost and resource utilization. Due to its substantial implications for the front end and back end of the cloud research business, resource utilization is undoubtedly an important factor that should be taken into consideration. When efficient resource usage is performed in the cloud, good load balancing is obtained. To address the preceding problems, we propose in this paper an innovative workflow scheduling technique in the cloud employing the Pathfinder algorithm (PFA) and the oppositional-based learning algorithm (OBL). The proposed scheduler focused on three objectives (i.e., total execution time, cost, and resource utilization). Simulation tests are carried out using the workflowsim toolkit to evaluate the effectiveness of the suggested approach. The results of the experiments proved that the proposed OBLPFA algorithm outperforms other algorithms called (Particle Swarm technique (PSO), dragonfly (DFA), the Arithmetic Optimization Technique (AOA), the Reptile Search technique (RSA), the Aquila Optimization method (AO), and Lion Optimization method (LOA)) in terms of total execution time, cost, and resource utilization.

A Tailored Particle Swarm and Egyptian Vulture Optimization-Based Synthetic Minority-Oversampling Technique for Class Imbalance Problem

Class imbalance is one of the significant challenges in classification problems. The uneven distribution of data samples in different classes may occur due to human error, improper/unguided collection of data samples, etc. The uneven distribution of class samples among classes may affect the classification accuracy of the developed model. The main motivation behind this study is the design and development of methodologies for handling class imbalance problems. In this study, a new variant of the synthetic minority oversampling technique (SMOTE) has been proposed with the hybridization of particle swarm optimization (PSO) and Egyptian vulture (EV). The proposed method has been termed SMOTE-PSOEV in this study. The proposed method generates an optimized set of synthetic samples from traditional SMOTE and augments the five datasets for verification and validation. The SMOTE-PSOEV is then compared with existing SMOTE variants, i.e., Tomek Link, Borderline SMOTE1, Borderline SMOTE2, Distance SMOTE, and ADASYN. After data augmentation to the minority classes, the performance of SMOTE-PSOEV has been evaluated using support vector machine (SVM), Naïve Bayes (NB), and k-nearest-neighbor (k-NN) classifiers. The results illustrate that the proposed models achieved higher accuracy than existing SMOTE variants.

ANALYSIS OF MACHINE LEARNING BASED RENEWABLE ENERGY SYSTEMS

The present research focuses on the development and optimization of the particle swarm techniques inspired in nature to predict wind speed in renewable energies with real-time wind farm data structures, of unique machine learning architectures for neural networks alongside certain mathematical and stochastic populations. The research work in this article includes six modules, i.e., designing proposed architectures of the neural network with variants based on population stochastic particle swarm (SPS) optimization and developed mathematical parameters in place of hidden neuron numbers to effectively predict the speed of renewable energy systems that reach the set number of neurons. The wind farm data sets are used for training, testing and validation of the proposed model of wind speed predictors. The final prediction model proposed involves the applicability of a neural wavelet network for a predictive wind velocity and the mother wavelet function is used to allow the hidden neurons and to measure the wind speed output with the reduction of the set parameters.

Autonomous Trajectory Optimisation for Intelligent Satellite Systems and Space Traffic Management

Recent advances in Artificial Intelligence (AI), sensing and computing technologies have led to the development of new promising concepts for the safe and efficient operation of Distributed Space Systems (DSS) in near-Earth orbits. The combined use of these technologies allows higher levels of autonomy in small satellite constellations or clusters, facilitating a more responsive and resilient approach to Space-Based Space Surveillance (SBSS), both in terms of data collection and data processing. This article presents an innovative DSS mission management approach exploiting multiple heterogeneous space platforms capable of autonomously calculating attitude and orbit raising manoeuvres to maximise mission efficiency and minimise the risk of collision with resident space objects. Furthermore, it addresses the development of reactive mission planning capabilities and lays foundations for introducing predictive system functionalities, by adopting heuristic multiple-direct-shooting trajectory optimisation algorithms based on the particle swarm technique. These functionalities provide DSS with higher levels of autonomy and support the introduction of new mission concepts while mitigating the threats of the space environment. The first verification case study addresses the problem of an orbit raising manoeuvre for debris collision avoidance, where the thrust activation and vectoring angle profiles are optimised to achieve the orbital semi-major axis increment which reduces the probability of collision to an acceptable level. An inverse-dynamic method is employed in the second case study, which addresses an optimal attitude reorientation problem for initial debris tracking. Typical attitude constraints and forbidden pointing zones are considered, highlighting the suitability and general applicability of the proposed DSS functionalities for the intended SBSS mission.

Optimización de parámetros en procesos de fundición a presión y compresión mediante el algoritmo de enjambre de partículas

In this article, an algorithm, based on the particle swarm technique (PSO), is developed to optimize die casting and compression casting processes, using mathematical models to describe the behavior of both processes. In compression casting the mathematical model describes a problem with multiple objectives and constraints, and in die casting the model describes a single objective problem with constraints. The development of the PSO algorithm was carried out with the FORTRAN 90 software, and the results were compared with those reported by a teaching-learning based optimization algorithm, (TLBO), demonstrating good capabilities in the optimization of parameters in die casting and by compression. It was observed that the PSO algorithm achieves an optimal solution in all processes and the computational time were minimal.

Modeling and optimization of reactive cotton dyeing using response surface methodology combined with artificial neural network and particle swarm techniques

Modeling and optimization of reactive cotton dyeing using response surface methodology combined with artificial neural network and particle swarm techniques

Novel Bio-Inspired Algorithm for Speed Control and Torque Ripple Reduction of Switched Reluctance Motor in Aerospace Application

Switched Reluctance Motor (SRM) is an electrical motor which operates by a reluctance torque that has stator and rotor salient poles. It also has a simple configuration and reasonable power electronics necessity for both AC and DC machines within adjustable-speed drives. The problem working on it is, SRM with doubly salient poles introduce a torque ripple in its output which leads to deteriorated performance on speed control. In this proposed work, torque ripple reduction and speed control of SRM is implemented, utilizing an asymmetrical converter with PI-PWM controller with the aid of Bio-inspired algorithms. By utilizing the best value of flux, torque ripple and integral square error of speed and settling times, controller design is performed. The proposed research work compares three optimization algorithms namely Crow Search Algorithm, an Improved Ant Lion Optimization Algorithm and Modified Chaotic Starling Particle Swarm technique to control the speed and torque reduction of SRM in aerospace applications. To obtain the optimal parameter values, optimization techniques are introduced and the optimized controller results are compared with other conventional controller results. In normal operation, the engine operates with two phases simultaneously, but it is modelled to meet the load specification, which is single and double phase faulty. Detection of fault gives a great outcome in control of an electrical drive system. Reduction of faulty operation time leads to better motor lifetime that is obtained by early detection of a fault. This paper presents how the five-phase switched reluctance motor is designed to meet the needs of flap actuators in mid-sized aircraft. Electrical systems are modelled to provide the right functionality at the right time in advanced aircraft. However, requirement of power for the landing gear and secondary flight control only a short duration. The system is powered on and off as required, thus preserving power. Control performance and fault analysis were verified by the results of the MATLAB simulation. Experiments have been carried out more effectively to compare the results obtained with those of simulations.

Adaptive Moth Flame Optimization based Load Shifting Technique for Demand Side Management in Smart Grid

The rising number of connected devices, energy management, and local generation facilities implies that the consumers have an increased responsibility to the energy sector. Demand-side management (DSM) techniques involve the consumer in assisting grid services, delay of capital expenses, and utility revenue streams through several behind the metre services. A variety of literature studies have presented many techniques for the implementation of DSM in smart grid systems. This study emphasizes the modelling of DSM utilizing a day-ahead load shifting approach as a minimization problem. The DSM is modelled as an optimization problem whose solution is attained through a nature-based adaptive moth flame (AMF) technique. The formulated work has been tested on three demand zones: residential, commercial, and industrial, with diverse controllable loads. A comparison of solutions based on reduced peak demand and operational costs is undertaken with the proposed AMF optimization algorithm. Finally, it is shown that the DSM technique based on the AMF technique exhibits better savings than the multi-agent and evolutionary approach in the residential and commercial sectors. However, the particle swarm technique proves a better alternative than the proposed technique in achieving cost savings.

Optimal PID Controller based on Convex Optimisation and Particle Swarm Techniques

This paper discusses the design and implementation of multiple optimisation algorithms for tuning a PID controller. A metaheuristic algorithm known as particle swarm optimisation (PSO) is used together with convex optimisation techniques, and the validity of the proposed algorithm is examined by comparing its performance with the performance of the classic PSO-PID and the well-known Ziegler–Nichols PID (ZN-PID). To obtain useful comparisons, a non-linear system, the air levitation system (AL), was utilised, controlled using a PID with three tuning strategies: the modified particle swarm technique (M-PSO-PID), the classic particle swarm (C-PSO-PID), and the Ziegler-Nichols method. The performance of the controllers was monitored through a cost function, the Integral Absolute Error (IAE). A disturbance was also imposed on the AL system to test the performance and the robustness of the algorithm under different conditions. The proposed algorithm is designed to determine and set the parameters for the M-PSO-PID controller utilising stabilising regions Kp, Ki, and Kd which form a convex problem. The PSO technique is then used to search for the optimal values inside the convex area. The simulation results for the control system show that the M-PSO-PID offered good closed-loop performance with advantages over other algorithms with regard to the system settling time and rise time. The results thus indicate the supremacy of the proposed algorithm.

An Approach for Optimal Feature Selection in Machine Learning using Global Sensitivity Analysis

The classification application is an important procedure for selecting the feature. The classification is mainly based on the features extracted from the object. You can select the best feature using the following three methods: wrapper selection, filter and embedded procedure. All three practices have been implemented by single or combined two approaches. As a result, there is no important feature in the classification process. This problem is solved by the proposed integrated global analysis of sensitivity. Each feature is selected in a classification based on the sensitivity of the feature and the correlation from the target vector in this integrated sensitivity and correlation approach. Likewise, the GSA approach uses a variety of filtering techniques for ranking attributes and optimization using particle swarm technique. Then, the optimum attributes are trained and tested using the Random Forest Classifier grid search via MATLAB software. In comparison to the existing method, wrapper-based selection, the performance of our integrated model is measured using sensitivity, specificity and accuracy. The experimental results of our proposed approach outweigh the sensitivities by 93.72%, 94.74% and the accuracy of 89.921% and 90% where, wrapper selection approach as sensitivity by 89.83% and the accuracy of 93%.

Gravity data interpretation by a two-sided fault-like geologic structure using the global particle swarm technique

We have developed a scheme for the interpretation of gravity data by a two-sided fault-like geological structure. The scheme is based on the particle swarm optimization and determines the model parameters of the upthrown and downthrown of the fault. It uses the second horizontal derivatives, which are calculated numerically from the measured gravity data by applying some filters of consecutive graticule spacings (the so-called here window lengths or s-values). The developed scheme has been verified on various noise-free synthetic models. Following that, the accuracy of the scheme has been assessed on a number of noisy synthetic examples. Finally, the developed technique has been successfully applied to two real data examples from India and the United States of America. It is found that the model parameters obtained from the scheme developed here are in good agreement with those reported in the published literature.