Standard Particle Swarm Optimization Research Articles

An experimental study on locating time-varying pollutant sources in a dynamic indoor environment with mechanical ventilation based on multi-robot cooperation

Using robots to locate indoor pollutant sources has attracted increasing attention in recent years, which is important for securing people's health and safety indoors. However, few studies have addressed the common scenarios for locating a time-varying source (release rate varies over time) in dynamic indoor environments with mechanical ventilation (condition of supply airflow varies over time). In such scenarios, the coupling effect of indoor dynamic airflow and time-varying source pose great challenges to robots locating the source. To address these challenges, in this study, a mobile robot olfaction system equipped with three robots was developed and 180 experiments in a laboratory were performed. Two methods, namely, the improved particle swarm optimization (IPSO) method and the improved whale swarm algorithm (IWOA) method, were first validated, followed by comparisons with two other methods, the standard particle swarm optimization (SPSO) method and wind utilization II (WUII) method. Both the IPSO and IWOA methods show potential for practical application due to their high success rates (averaging 78.33% and 88.33%, respectively) and localization efficiency, while the success rates of the SPSO and WUII methods (less than 50%) are insufficient for practical application.

Kinematic parameters calibration of industrial robot based on RWS-PSO algorithm

The positioning accuracy of an industrial robot has a significant impact on its application in precision manufacturing, and it is necessary to calibrate robot kinematic parameters. Previous studies have established numerous nonlinear equations to solve the kinematic parameters, which are complicated and time consuming. A standard particle swarm optimization (PSO) algorithm is limited by long running time and low solution efficiency. Therefore, in this study, a dynamic particle swarm optimization algorithm based on roulette wheel selection (RWS-PSO) is proposed to realize the kinematic parameters calibration. First, a kinematics model is constructed using the standard Denavit-Hartenberg (D-H) method, and the theoretical and actual values of the spatial position of the robot endpoint are obtained via forward kinematics and a Laser Tracker, respectively. Next, the kinematic parameters calibration problem is transformed into a solution of a high-dimensional nonlinear equation using the proposed RWS-PSO algorithm. In the proposed RWS-PSO algorithm, the inertia factor is considered as linearly decreasing and the number of particles is selected by the roulette wheel selection (RWS) to improve its computational efficiency. The proposed RWS-PSO and standard PSO algorithms are compared based on various indices by simulation. The results reveal that the time cost of the RWS-PSO algorithm is much lower than that of the standard PSO algorithm on the basis of high precision and a reduced running time of approximately 53%. Finally, the kinematic parameter errors obtained by the two algorithms are compensated. According to the experimental results, the positioning accuracy of the robot in three ( x-, y-, and z-) directions is improved by 78%, 46%, and 67%, respectively, compared to that of before compensation, which proves that the RWS-PSO algorithm is effective and practical for kinematic parameters calibration.

Effects of flight parameters for plant protection UAV on droplets deposition rate based on a 3D simulation approach

With the development of aviation agricultural technology, the number of farmers adopting the use of drones in daily agricultural activities is growing rapidly in recent decades. Among these, a large portion constitutes agricultural drones being used in pest control and crop protection practices, e.g. agriculture spraying of pesticides. Spraying pesticides with drones have proven to be faster than other traditional methods. On the downside, flight time and range of Unmanned Aerial Vehicles (UAV) are often limited. Thus, a proper arrangement of flight height and velocity will greatly improve spraying efficiency. A new strategy to optimize the flight parameters, i.e. flight height and flight velocity, for fixed-wing UAV with a 3D simulation-based approach together with an automatic optimization algorithm was proposed in this study. To find the optimal parameters for a UAV to fly and spray under certain environmental conditions, a three-dimensional model of the target crop was established first, followed by a detailed simulation of droplet spraying. As a demonstration case, a grass model was developed and used as the target plant, and a physics-based method was used to simulate realistically the movement of the droplets in the air as well as the interaction between the droplets and the plant to obtain the droplet deposition rate under the specified operating parameters. Furthermore, the standard Particle Swarm Optimization (PSO) algorithm was used to optimize the UAV operating parameters to obtain the best operating parameters. The results indicate that using the standard PSO algorithm to optimize the operating parameters of the drone could significantly improve the deposition rate and find the best operating parameters. Keywords: unmanned aerial vehicle, droplet deposition, 3D simulation, particle swarm optimization algorithm DOI: 10.25165/j.ijabe.20231601.6581 Citation: Xu L F, Yang Z Z, Huang Z S, Ding W L, Buck-Sorlin G. Effects of flight parameters for plant protection UAV on droplets deposition rate based on a 3D simulation approach. Int J Agric & Biol Eng, 2023; 16(1): 66–72.

Improved Adaptive PSO-based Gain Tuning for PID Controllers in Utility Boilers

In this paper, the stabilization and set point tracking of the utility boiler is achieved via the Proportional-Integral-Derivative (PID) controller, tuned using the versions of metaheuristic Particle Swarm Optimization (PSO) algorithms. The utility boiler model dynamics are based on the Bell and Astrom boiler turbine system. Initially, the non-linear mathematical model is obtained using the first principles data from the plant model. Further, Taylor series approximation is applied to linearize the non-linear equations around the equilibrium point to implement a stable controller for the plant. Finally, the PID controller gains for the plant model are determined using the PSO algorithms to achieve the desired response. For tuning the PID controller, three PSO algorithms namely standard PSO, Adaptive PSO (APSO) and Improved Adaptive PSO (IAPSO) are proposed. The performance of the proposed PSO based PID control algorithms are evaluated using error metrics such as Integral Absolute Error (IAE), Integral Time Absolute Error (ITAE) and Integral Square Error (ISE). The simulation results show that the proposed APSO and IAPSO algorithms have improved performance when compared to regular PSO algorithms and are capable of optimizing the control parameters.

Modified PSO tuned two stage lead lag STATCOM control action for small signal stability enhancement of variable wind integrated power system

In this work Modified PSO (MPSO) is proposed to tune two stage lead lag control action based on STATCOM for enhancing small signal stability of an intermittent wind source penetrated power system. The Modified PSO has been compared with standard PSO and fractional PSO to tune controller parameters of STATCOM based control action. Multi machine system subjected to sudden and intermittent wind variations have been considered in this work. Subject to this wind penetration the lead lag control action with only PSO and two stage lead lag control action with fractional PSO and modified PSO has been implemented to damp the system oscillations pertaining to initial operating condition. It was found that subject to intermittent nature of wind penetration the rotor angle small signal stability is being hampered and subject to this when the modified PSO tuned two stage lead lag STATCOM control action is applied, it damps the oscillations heavily and the results have been compared with only PSO and fractional PSO control algorithms.

Improved Particle Swarm Optimization for Laser Cutting Path Planning

This research focuses on the long empty cutting path problem during the laser cutting process by employing an improved proximity method to establish the starting point set in complex closed graphics. Specifically, this work improves the particle swarm algorithm and proposes the Levy Flight, power function, and Singer map employed particle swarm optimization (LPSPSO) to avoid the disadvantages of the standard particle swarm optimization (PSO) algorithm. Specifically, the comprehensive prospect-regret theoretical model evaluation value is used as the fitness value to guide the algorithm’s evolution and adaptively adjust the parameters in the LPSPSO algorithm, including the inertia weight power function, the learning factors, and the chaotic random number based on the Singer chaotic map. Additionally, the Levy flight is introduced to disturb the particles and prevent local optimization. This is achieved by adjusting the Levy flight threshold based on the distance between the particles to prevent the Levy flight from starting prematurely and increasing the calculation burden. To verify the performance of the LPSPSO algorithm, it was challenged against three state-of-the-art algorithms on 22 benchmark test instances and a laser cutting problem, with the results revealing that the LPSPSO algorithm has a better performance and can be used to solve the empty length of the laser cutting path problem.

An improved particle swarm optimization combined with double-chaos search.

Particle swarm optimization (PSO) has been successfully applied to various complex optimization problems due to its simplicity and efficiency. However, the update strategy of the standard PSO algorithm is to learn from the global best particle, making it difficult to maintain diversity in the population and prone to premature convergence due to being trapped in local optima. Chaos search mechanism is an optimization technique based on chaotic dynamics, which utilizes the randomness and nonlinearity of a chaotic system for global search and can escape from local optima. To overcome the limitations of PSO, an improved particle swarm optimization combined with double-chaos search (DCS-PSO) is proposed in this paper. In DCS-PSO, we first introduce double-chaos search mechanism to narrow the search space, which enables PSO to focus on the neighborhood of the optimal solution and reduces the probability that the swarm gets trapped into a local optimum. Second, to enhance the population diversity, the logistic map is employed to perform a global search in the narrowed search space and the best solution found by both the logistic and population search guides the population to converge. Experimental results show that DCS-PSO can effectively narrow the search space and has better convergence accuracy and speed in most cases.

WACPN: A Neural Network for Pneumonia Diagnosis.

Community-acquired pneumonia (CAP) is considered a sort of pneumonia developed outside hospitals and clinics. To diagnose community-acquired pneumonia (CAP) more efficiently, we proposed a novel neural network model. We introduce the 2-dimensional wavelet entropy (2d-WE) layer and an adaptive chaotic particle swarm optimization (ACP) algorithm to train the feed-forward neural network. The ACP uses adaptive inertia weight factor (AIWF) and Rossler attractor (RA) to improve the performance of standard particle swarm optimization. The final combined model is named WE-layer ACP-based network (WACPN), which attains a sensitivity of 91.87±1.37%, a specificity of 90.70±1.19%, a precision of 91.01±1.12%, an accuracy of 91.29±1.09%, F1 score of 91.43±1.09%, an MCC of 82.59±2.19%, and an FMI of 91.44±1.09%. The AUC of this WACPN model is 0.9577. We find that the maximum deposition level chosen as four can obtain the best result. Experiments demonstrate the effectiveness of both AIWF and RA. Finally, this proposed WACPN is efficient in diagnosing CAP and superior to six state-of-the-art models. Our model will be distributed to the cloud computing environment.

Modified Particle Swarm Optimization Based on Aging Leaders and Challengers Model for Task Scheduling in Cloud Computing

In cloud computing, a shared, configurable pool of computing resources is made accessible online and allocated to users according to their needs. It is essential for a cloud provider to schedule jobs in the cloud to keep up service quality and boost system efficiency. In this paper, we present a scheduling technique based on modified particle swarm optimization to combat the issues of excessively long scheduling time and high computation costs associated with scheduling jobs in a cloud environment. The modified PSO is used to allocate the jobs to virtual machines in order to minimize the objective function consisting of cost and makespan. The algorithm relies on biological changes that occur in organisms to regulate premature convergence and improve local search capability. The technique is analyzed and simulated using CloudSim, and the simulation results demonstrate that the proposed approach decreases makespan and cost effectively as compared to standard PSO.

Parameter estimation based on novel enhanced self‐learning particle swarm optimization algorithm with Levy flight for PMSG

AbstractA novel parameter estimation method is proposed for the permanent magnet synchronous generator (PMSG), which is implemented by an enhanced self‐learning particle swarm optimization algorithm with Levy flight (SLPSO), and the problem of lower parameter estimation precision of standard PSO is obviated. This method injects currents of different intensities into the d‐axis in a time‐sharing manner to solve the problem of equation under‐ranking, and the mathematical model for full‐rank parameter estimation is developed. The speed term of PSO is simplified to expedite the convergence of PSO, and a strategy with chaotic decline for the inertia weight of PSO is adopted to strengthen its ability to jump out of the local optimum. Moreover, the self‐learning dense fleeing strategy (SLDF) is proposed where particles perform diffusion learning based on population density information and Levy flight, the evolutionary unitary problem and human intervention in the evolutionary process is averted. Furthermore, the memory tempering annealing algorithm (MTA) and greedy algorithm (GA) is integrated into the algorithm, MTA can facilitate the exploration of potentially better regions, and GA for local optimization enhances the convergence speed and accuracy in late stage of the algorithm. Comparing the proposed method with several existing PSO algorithms through simulation and experiments, the experimental data show that the proposed method can effectively track variable parameters under different working conditions and has better robustness.

Magnetic anomaly inversion through the novel barnacles mating optimization algorithm

Dealing with the ill-posed and non-unique nature of the non-linear geophysical inverse problem via local optimizers requires the use of some regularization methods, constraints, and prior information about the Earth's complex interior. Another difficulty is that the success of local search algorithms depends on a well-designed initial model located close to the parameter set providing the global minimum. On the other hand, global optimization and metaheuristic algorithms that have the ability to scan almost the entire model space do not need an assertive initial model. Thus, these approaches are increasingly incorporated into parameter estimation studies and are also gaining more popularity in the geophysical community. In this study we present the Barnacles Mating Optimizer (BMO), a recently proposed global optimizer motivated by the special mating behavior of barnacles, to interpret magnetic anomalies. This is the first example in the literature of BMO application to a geophysical inverse problem. After performing modal analyses and parameter tuning processes, BMO has been tested on simulated magnetic anomalies generated from hypothetical models and subsequently applied to three real anomalies that are chromite deposit, uranium deposit and Mesozoic dike. A second moving average (SMA) scheme to eliminate regional anomalies from observed anomalies has been examined and certified. Post-inversion uncertainty assessment analyses have been also implemented to understand the reliability of the solutions achieved. Moreover, BMO’s solutions for convergence rate, stability, robustness and accuracy have been compared with the solutions of the commonly used standard Particle Swarm Optimization (sPSO) algorithm. The results have shown that the BMO algorithm can scan the model parameter space more extensively without affecting its ability to consistently approach the unique global minimum in this presented inverse problem. We, therefore, recommend the use of competitive BMO in model parameter estimation studies performed with other geophysical methods.

Workflow Scheduling in Cloud Computing Environment by Combining Particle Swarm Optimization and Grey Wolf Optimization

Scheduling workflows is a vital challenge in cloud computing due to its NP-complete nature and if an efficient workflow task scheduling algorithm is not used then it affects the system’s overall performance. Therefore, there is a need for an efficient workflow task scheduling algorithm that can distribute dependent tasks to virtual machines efficiently. In this paper, a hybrid workflow task scheduling algorithm based on a combination of Particle Swarm Optimization and Grey Wolf Optimization (PSO GWO) algorithms, is proposed. PSO GWO overcomes the disadvantages of both PSO and GWO algorithms by improving the exploitation (local search) of PSO algorithm and exploration (global search) of GWO algorithm. This leads to better balance between exploration and exploitation, consequently it minimizes the makespan with 5.52% compared to GWO and 3.68% compared to PSO. The degree of imbalance reduced upto 33.22% compared to GWO and 17.61% compared to PSO, improves the convergence rate as well depending on number tasks and iterations. CloudSim tool is used to evaluate the proposed algorithm. The simulation results confirmed that the proposed method performs better than both of the standard PSO and GWO in terms of makespan, degree of imbalance and convergence rate

Online measurement method for roundness of large aluminum alloy ring based on laser ranging principle

Aluminum alloy rings are widely used in aerospace, energy, petrochemical, and equipment applications. The loss of roundness often occurs because of the weak rigidity and unstable plasticity of aluminum alloy rings. Therefore, an online measurement technique for the roundness of aluminum alloy rings is critical for ensuring the processing quality of aluminum alloy rings. In this paper, an online roundness measurement method based on the laser ranging principle and improved particle swarm optimization (IPSO) is proposed with the aid of the rotation of the ring during the rolling process. First, the distance from the outer surface of the ring to the laser rangefinder was measured using a triangular laser rangefinder with the help of the rotation of the ring in the rolling process. Second, the distance data were denoized using an improved Kalman filter algorithm to reduce the influence of random noise on the measurement results. Finally, the roundness and geometric center were evaluated using the IPSO algorithm combined with the minimum zone circle (MZC) method and compared with the results obtained using the standard particle swarm optimization (PSO) algorithm, genetic algorithm (GA), and least square (LS) algorithm. The experimental results show that the roundness measured by the IPSO algorithm is less than 0.06 mm, demonstrating better performance in the roundness evaluation. The measurement method for roundness proposed in this paper is suitable for the online evaluation of large aluminum alloy rings, avoiding the influence of ring size on roundness measurement.

Cooling performance enhancement study of single droplet impingement on heated hole-patterned surfaces using improved GQPSO algorithm

A droplet impingement on solid surfaces plays a critical role in various spray and heat transfer applications. The main objective of this study is to investigate how to enhance the cooling performance of the single droplet impingement process on the heated hole-patterned surfaces by finding desired conditions using optimization techniques. The cooling effectiveness and total contact area at the maximum droplet spread were chosen for objective functions to be maximized. To estimate the optimal values of the unknown variables efficiently, the improved Gaussian-behaved particle swarm optimization (GQPSO) algorithm with a modified local attractor using Gaussian distributed random numbers was newly proposed and applied to the single droplet impingement problem. To verify the improved search performance of the proposed GQPSO approach, the optimization results were compared with the standard particle swarm optimization (PSO) and original GQPSO methods with the same conditions. By comparing the estimated maximum values of given objective functions, calculation times and numbers of penalty particles, we verified that the proposed GQPSO approach could search for more optimal solutions quickly and efficiently due to its improved local fine-tuning capability with the modified local attractor.

An Improved Particle Swarm Optimization Algorithm for Data Classification

Optimisation-based methods are enormously used in the field of data classification. Particle Swarm Optimization (PSO) is a metaheuristic algorithm based on swarm intelligence, widely used to solve global optimisation problems throughout the real world. The main problem PSO faces is premature convergence due to lack of diversity, and it is usually stuck in local minima when dealing with complex real-world problems. In meta-heuristic algorithms, population initialisation is an important factor affecting population diversity and convergence speed. In this study, we propose an improved PSO algorithm variant that enhances convergence speed and population diversity by applying pseudo-random sequences and opposite rank inertia weights instead of using random distributions for initialisation. This paper also presents a novel initialisation population method using a quasi-random sequence (Faure) to create the initialisation of the swarm, and through the opposition-based method, an opposite swarm is generated. We proposed an opposition rank-based inertia weight approach to adjust the inertia weights of particles to increase the performance of the standard PSO. The proposed algorithm (ORIW-PSO-F) has been tested to optimise the weight of the feed-forward neural network for fifteen data sets taken from UCI. The proposed techniques’ experiment result depicts much better performance than other existing techniques.

An Efficient Hybrid Particle Swarm and Teaching-Learning-Based Optimization for Arch-Dam Shape Design

Particle swarm optimization is a popular meta-heuristic with highly explorative features; however, in its standard form it suffers from a poor convergence rate and weak search refinement on multi-dimensional problems. The present work improves the conventional particle swarm optimizer in three ways: adding a greedy selection for better intensification; embedding an extra movement borrowed from teacher–learner-based optimization; and utilizing a neighborhood strategy by averaging over a random half of the swarm. The performance of the proposed method is subsequently evaluated on three sets of problems. The first set includes uni-modal, multi-model, separable and non-separable test functions. The proposed method is compared with a standard particle swarm optimizer and its variants as well as other meta-heuristic algorithms. Engineering benchmark problems including the optimal design of a tubular column, a coiled spring, a pressure vessel and a cantilever beam constitute the second set. The third set includes constrained sizing design of a 120-bar dome truss and the optimal shape design of the Morrow Point double-arch concrete dam as a practical case study. Numerical results reveal considerable enhancement of the standard particle swarm via the proposed method to exhibit competitive performance with the other studied meta-heuristics. In the optimal design of Morrow Point Dam, the proposed method resulted in a material consumption 21 times smaller than the best of the initial population and 26% better than a recommended practical design.

Autonomous vehicle path planning for smart logistics mobile applications based on modified heuristic algorithm

In this study, a heuristic algorithm is used to find an optimal route for smart logistics loading and unloading applications. Various environments, such as traditional building blocks, satellite images, terrain environments, and Google map environments are developed by converting into a binary occupancy grid and used to optimize the viable path in the smart mobile logistics application. The proposed autonomous vehicle (AV) route planning navigation approach is to forecast the AV’s path until it detects an imminent obstacle, at which point it should turn to the safest area before continuing on its route. To demonstrate the path navigation results of proposed algorithms, a navigational model is developed in the MATLAB/Simulink 2D virtual environment. The particle swarm optimization (PSO) method, the Bat search algorithm, and its proposed variants are used to identify a smooth and violation-free path for a given application environment. The proposed variants improve the algorithm’s effectiveness in finding a violation-free path while requiring less time complexity by using cubic spline curve interpolation and its improved constriction factor. Extensive simulation and benchmark validation results show the proposed standard PSO has a significantly shorter violation-free path, quick convergence rate and takes less time to compute the distance between loading and unloading environment locations than the cooperative coevolving PSO, Bat algorithm, or modified frequency Bat algorithms.

Optimizing an adaptive fuzzy logic controller of a 3-DOF helicopter with a modified PSO algorithm

This paper investigates the controller optimization for a helicopter system with three degrees of freedom (3-DOF). The system is extensively nonlinear and highly sensitive to the controller’s parameters, making it a real challenge to study these parameters’ effects on the controller’s performance. We combined fuzzy logic with adaptive control theory to control the system and used metaheuristic algorithms to determine these parameters. Then, we compare the results with the controller optimized through the standard PSO and PID controller. The results indicate the high ability of MPSO to perform the global search and to find a reasonable search space. The proposed method’s effectiveness and robustness properties are shown through computer simulations, while the system is subject to uncertainties and disturbance. We also prove the efficiency of the MPSO algorithm by comparing it with the standard PSO and six other well-known metaheuristic algorithms and analyzing the results by statistical tests.

Dynamic Sub-Swarm Approach of PSO Algorithms for Text Document Clustering.

Text document clustering is one of the data mining techniques used in many real-world applications such as information retrieval from IoT Sensors data, duplicate content detection, and document organization. Swarm intelligence (SI) algorithms are suitable for solving complex text document clustering problems compared to traditional clustering algorithms. The previous studies show that in SI algorithms, particle swarm optimization (PSO) provides an effective solution to text document clustering problems. This PSO still needs to be improved to avoid the problems such as premature convergence to local optima. In this paper, an approach called dynamic sub-swarm of PSO (subswarm-PSO) is proposed to improve the results of PSO for text document clustering problems and avoid the local optimum by improving the global search capabilities of PSO. The results of this proposed approach were compared with the standard PSO algorithm and K-means algorithm. As for performance assurance, the evaluation metric purity is used with six benchmark data sets. The experimental results of this study show that our proposed subswarm-PSO algorithm performs best with high purity comparing the standard PSO and K-means traditional algorithms and also the execution time of subswarm-PSO comparatively takes a little less than the standard PSO algorithm.

A Novel Blind Image Source Separation Using Hybrid Firefly Particle Swarm Optimization Algorithm

Signal and image separation are extensively used in numerous imaging applications and communication systems. In this paper, a novel Blind Source Separation (BSS) approach, based on the Hybrid Firefly Particle Swarm Optimization (HFPSO), is proposed for separating mixed images. This approach processes the observed source without any prior knowledge about the model and the statistics of the source signal. The proposed method presents high robustness against local minima and converges quickly to the global minimum. Via numerical simulations, the proposed approach is tested and validated in comparison with standard Particle Swarm Optimization (PSO), Robust Independent Component Analysis (RobustICA), and Artificial Bee Colony (ABC) algorithms. The obtained results show that the presented technique outperforms the existing ones in terms of quality of image separation, the Signal-to-Noise Ratio (PSNR), and Structural Similarity Index Measure (SSIM). Moreover, the obtained results demonstrate that our approach provides also promising results in image separation from noisy mixtures.