Improved Particle Swarm Optimization Research Articles

TOA-based direct localization in shallow water multipath environments: CRLB analysis and optimal sensor deployment

Existing research explores the optimal sensing deployment in line-of-sight conditions, leveraging Cramér-Rao Lower Bound (CRLB) for optimization analysis. However, determining an optimal sensing deployment for direct localization in shallow water, characterized by distinctive multipath effects, poses a more significant challenge. This paper studies the optimal sensor deployment for direct localization grounded on the Time of Arrival (TOA) in a shallow water multipath environment. To address this problem, an initial derivation of the CRLB for accurate source localization is conducted, bearing in mind a multipath signal loss model. Subsequently, an optimization problem is articulated to minimize the trace of CRLB while adhering to the angular and range constraints to ascertain optimal sensor placement. Two necessary conditions for the objective function to attain a theoretical minimum are drawn out to find the closed-form solution. Nevertheless, it is impractical to satisfy these conditions concurrently under multipath circumstances. To swiftly attain the (sub-)optimal solution, an improved Particle Swarm Optimization (IPSO) algorithm is proposed, employing the previously derived conditions to initialize and update the inertia weights. Extensive simulation results based on the BELLHOP toolbox corroborate the theoretical analysis and the superior performance of the suggested sensor deployment scheme.

Deep learning approach for brain tumor classification using metaheuristic optimization with gene expression data

AbstractThis study addresses the critical challenge of accurately classifying brain tumors using artificial intelligence. Early detection is crucial, as untreated tumors can be fatal. Despite advances in AI, accurately classifying tumors remains a challenging task. To address this challenge, we propose a novel optimization approach called PSCS combined with deep learning for brain tumor classification. PSCS optimizes the classification process by improving Particle Swarm Optimization (PSO) exploitation using Cuckoo search (CS) algorithm. Next, classified gene expression data of brain tumor using Deep Learning (DL) to identify different groups or classes related to a particular tumor along with the PSCS optimization technique. The proposed optimization technique with DL achieves much better classification accuracy than other existing DL and Machine learning models with the different evaluation matrices such as Recall, Precision, F1‐Score, and confusion matrix. This research contributes to AI‐driven brain tumor diagnosis and classification, offering a promising solution for improved patient outcomes.

Improved Particle Swarm Optimization With Less Manual Intervention for Photonic Inverse Design

Improved Particle Swarm Optimization With Less Manual Intervention for Photonic Inverse Design

Three-Dimensional Iterative Enhancement for Coverage Hole Recovery in Underwater Wireless Sensor Networks

The efficient coverage of underwater wireless sensor networks (UWSNs) has become increasingly important because of the scarcity of underwater node resources. Complex underwater environments, water flow forces, and undulating seabed reduce the coverage effect of underwater nodes, even leading to coverage holes in UWSNs. To solve the problems of uneven coverage distribution and coverage holes, a three-dimensional iterative enhancement algorithm is proposed for UWSN coverage hole recovery using intelligent search followed by virtual force. Benefiting from biological heuristic search algorithms, improved particle swarm optimization is applied for node pre-coverage. With the change in iteration times, the adaptive inertia weight, acceleration factor, and node position are constantly updated. To avoid excessive coverage holes caused by search falling into local optimum, underwater nodes are considered as particles in the potential field whose virtual forces are calculated to guide nodes towards higher coverage positions. In addition, based on the optimal node location obtained by the proposed algorithm, the monitoring area is divided based on the clustering idea. The underwater routing protocol DBR based on depth information is subsequently used to optimize node residual energy, and its average is calculated comprehensively and compared with the other three coverage algorithms using the DBR routing protocol. Based on the experimental data, after 100 iterations, the coverage rates for BES, 3D-IVFA, DABVF, and the proposed algorithm are 83.28%, 88.85%, 89.31%, and 91.36%, respectively. Moreover, the proposed algorithm is further verified from the aspects of different node numbers, coverage efficiency, node movement trajectory, coverage hole, and average residual energy of nodes, which provides conditions for resource development and scientific research in marine environments.

Non-parametric dynamics modeling for unmanned surface vehicle using spectral metric multi-output Gaussian processes learning

High-precision dynamic models are crucial for various marine missions and the control and navigation of unmanned surface vehicles. However, the nonlinear, strong coupling, complex structure of these vehicles poses challenges in building such models. Traditional single-output models have limitations in data utilization and high computational complexity. Therefore, this paper proposes a spectral metric multi-output Gaussian process for continuous-time dynamic modeling of unmanned surface vehicles. Specifically, a spectral covariance matrix is constructed in this scheme to establish a spectral metric multi-output Gaussian processes learning model that captures correlations under different degrees of freedom. Meanwhile, an improved particle swarm optimization algorithm is used to dynamically tuning the correlation matrix for improved model accuracy, and utilizing a subspace index to enhance computational efficiency in multi-output Gaussian processes learning. Moreover, the scheme can conduct real-time simulation under environmental disturbances. Finally, the proposed scheme is applied to an unmanned surface vehicle, and the results demonstrate that the scheme is an effective modeling tool for multi degree-of-freedom coupled dynamics.

Retracted: Low-Voltage Diagnosis of Energy Distribution Network Based on Improved Particle Swarm Optimization Algorithm

Retracted: Low-Voltage Diagnosis of Energy Distribution Network Based on Improved Particle Swarm Optimization Algorithm

Enhancing PV Systems with Intelligent MPPT and Improved control strategy of Z-Source Inverter

Abstract The Improved Z-Source Inverter (IZSI) has gained attention in the photovoltaic industry for its ability to boost PV voltage with a single-stage topology, simplifying system design and reducing costs. However, research on integrating IZSI into PV systems, particularly regarding the Maximum Power Point Tracker (MPPT) and IZSI control strategy, is limited. This study proposes an Intelligent Improved Particle Swarm Optimization (IPSO) algorithm as an MPPT method for PV systems under constant and varying irradiance conditions. The IPSO algorithm is compared to the FPA, CSA, and traditional MPPT algorithm (PSO), and the results demonstrate that IPSO outperforms all algorithms in terms of speed, efficiency, and convergence in finding the Maximum Power Point (MPP). Two methods, Simple Boost Control (SBC) and Maximum Constant Boost Control with Third Harmonic Injection (THIMCBC), are employed to control IZSI. Simulation results using MATLAB-Simulink show that both strategies successfully find and track the MPP, but THIMCBC exhibits superior voltage-boosting performance compared to SBC. Overall, the proposed IZSI topology with the IPSO MPPT method and THIMCBC IZSI control strategy offers several advantages, including improved voltage boost ability, reduced z-source capacitor voltage stress, inherent inrush current limitation, and cost-effectiveness. These advantages make the proposed system a promising solution for photovoltaic systems.

Step-like displacement prediction and failure mechanism analysis of slow-moving reservoir landslide

Landslides triggered by extreme rainfall due to global climate change are becoming more frequent. The Earth surface processes activity and landform evolution caused by landslide movement seriously affect human survival and the environment. In recent years, the reservoir slope located in the Three Gorges reservoir (TGR) area has been subject to the combined effects of extreme rainfall and reservoir regulation, which have been highly susceptible to inducing landslides. Reservoir landslides usually show step-like slow movement characteristics. The causes of slow-moving landslides are complex, the mechanisms behind them are difficult to grasp, and it may evolve into violent landslide disasters. Therefore, accurately predicting step-like displacement in slow-moving landslides is an effective solution for risk reduction. To achieve the accurate prediction of the step-like displacement of slow-moving landslides and the analysis of the failure mechanism, this study establishes a prediction model applicable to step-like displacement prediction of slow-moving reservoir landslides is developed based on 16 years of continuously monitored cumulative displacement of the Jiuxianping landslide, combined with the Empirical Mode Decomposition (EMD) method and the Improved Particle Swarm Optimization (IPSO) optimized the Long short-term memory (LSTM) neural network. This approach can effectively improve step-like displacement prediction accuracy. In particular, this approach can better represent the association between the step displacement of slow-moving landslides with environmental characteristics (e.g., rainfall, reservoir level). The study reveals that there are differences in the mechanisms and influencing conditions that cause the deformation of different areas of the landslide. Additionally, the reservoir water level fluctuation and seasonal rainfall are identified as the primary reasons contributing to the stepped displacement of the Jiuxianping landslide. The method provides a basis for the step-like displacement prediction, instability mechanism and landform evolution study of slow-moving reservoir landslides and provides valuable guidance for the prevention and control of similar reservoir geomorphic hazards worldwide.

Dynamic parameter identification based on improved particle swarm optimization and comprehensive excitation trajectory for 6R robotic arm

PurposeRobotic arms’ interactions with the external environment are growing more intricate, demanding higher control precision. This study aims to enhance control precision by establishing a dynamic model through the identification of the dynamic parameters of a self-designed robotic arm.Design/methodology/approachThis study proposes an improved particle swarm optimization (IPSO) method for parameter identification, which comprehensively improves particle initialization diversity, dynamic adjustment of inertia weight, dynamic adjustment of local and global learning factors and global search capabilities. To reduce the number of particles and improve identification accuracy, a step-by-step dynamic parameter identification method was also proposed. Simultaneously, to fully unleash the dynamic characteristics of a robotic arm, and satisfy boundary conditions, a combination of high-order differentiable natural exponential functions and traditional Fourier series is used to develop an excitation trajectory. Finally, an arbitrary verification trajectory was planned using the IPSO to verify the accuracy of the dynamical parameter identification.FindingsExperiments conducted on a self-designed robotic arm validate the proposed parameter identification method. By comparing it with IPSO1, IPSO2, IPSOd and least-square algorithms using the criteria of torque error and root mean square for each joint, the superiority of the IPSO algorithm in parameter identification becomes evident. In this case, the dynamic parameter results of each link are significantly improved.Originality/valueA new parameter identification model was proposed and validated. Based on the experimental results, the stability of the identification results was improved, providing more accurate parameter identification for further applications.

Application of the improved particle swarm optimization method in slope probability analysis

To improve the probability analysis efficiency of slope, this paper proposes an improved particle swarm optimization (IPSO) method to determine the critical sliding surface (CSS) and its corresponding minimum safety factor, which is used to calculate the failure probability of slope. Based on the random fields generated by using Karhunen-Loève (KL) expansion method, the simplified Bishop’s method combined with the entry and exit method, the particle swarm optimization (PSO) method, and the IPSO method is used to determine the CSS and its corresponding minimum factor of safety. Then, Monte Carlo simulation is used to estimate the failure probability of slope. The application potential of the IPSO method is demonstrated by re-analyzing two examples. Meaningful comparisons are made to demonstrate the calculating accuracy and calculating efficiency of the IPSO method in searching for the minimum safety factor of slope. Results show that the IPSO can accurately and efficiently determine the minimum safety factor in slope probability analysis considering the spatially variable soils. The IPSO method provides a promising tool for an efficient slope probability analysis.

Filament-necking localization method via combining improved PSO with rotated rectangle algorithm for safflower-picking robots

Safflower is one of the most important specialty economic crops in the world. Safflower blossoms are harvested continuously for 3–5 times. Untimely picking can affect the opening of filaments in the next crop, resulting in reduced filament production and serious economic losses. However, the safflower images collected by picking robots were affected by lighting conditions or complex backgrounds. The existing segmentation algorithms have the problem of insufficient segmentation or over-segmentation, because of the small safflower size and localization area. Therefore, combining improved particle swarm optimization (PSO) with a rotated rectangle algorithm based on the filament-necking localization method for safflower-picking robots was proposed. The R-component image in RGB color space was extracted as a preprocessing sample by combining the color features of safflower. The inertia weights in the PSO algorithm are improved to enhance the performance of the algorithm. An adaptive nonlinear function is introduced to search for the optimal threshold and initially segmented to obtain the binary image. Then, the barycenter and minimum outer rectangle of the contour were set up using the rotated rectangle algorithm based on the geometric features of the filaments. The circular region of interest (Cir-ROI) of filament-necking is determined. The Zhang-Suen refinement algorithm was used for skeleton extraction to design an algorithm for localizing the picking point of safflower filaments. To test safflower images collected in complex environments, the results of average processing time were 0.14 s, and the average relative target area error rate was 19.33 %. Moreover, the localization accuracy of the picking point was 89.75 %. The filament-necking localization method provides a theoretical basis and experimental data support for damage reduction and efficient harvesting of safflower filaments.

Efficient and precise docking trajectory optimization for the ship block assembly

The assembly of the ship block is an extremely important stage of the shipbuilding process. Nevertheless, currently, the manual assembly efficiency is low, the accuracy is poor, and collision is very easy to occur. Therefore, there is an urgent need to conduct technical research on the automatic docking of ship blocks. The core of the automated docking technology is the attitude estimation and the trajectory planning of the posturing equipment. However, current data measurement and point set matching methods lead to large attitude-estimation errors, and it is difficult to meet the accuracy requirements of the assembly. Moreover, the current ship block trajectory planning methods pay more attention to single metrics, for example, time or energy consumption, while omitting the shock degree. In response to the above, this study first proposes a high-precision matching method for measuring point sets, in order to estimate the exact attitude of the ship block. Subsequently, trajectory translation for the block is performed using the seventh-degree polynomial. On this basis, a nonlinear weighted improved particle swarm optimization (IPSO) method is proposed to optimize the time, energy consumption and shock degree in the ship block trajectory planning process. Finally, the accuracy of the matching optimization is validated by simulation analysis and it is concluded that the seventh-degree polynomial leads to less shock than other polynomials. Furthermore, the shock force does not change abruptly even when the ship block is poised in steps. Through IPSO, the energy consumption and shock degree performance indices are optimized by 37.07% and 50.06%, respectively, in the ship block translation process.

AUV Collision Avoidance Planning Method Based on Deep Deterministic Policy Gradient

Collision avoidance planning has always been a hot and important issue in the field of unmanned aircraft research. In this article, we describe an online collision avoidance planning algorithm for autonomous underwater vehicle (AUV) autonomous navigation, which relies on its own active sonar sensor to detect obstacles. The improved particle swarm optimization (I-PSO) algorithm is used to complete the path planning of the AUV under the known environment, and we use it as a benchmark to improve the fitness function and inertia weight of the algorithm. Traditional path-planning algorithms rely on accurate environment maps, where re-adapting the generated path can be highly demanding in terms of computational cost. We propose a deep reinforcement learning (DRL) algorithm based on collision avoidance tasks. The algorithm discussed in this paper takes into account the relative position of the target point and the rate of heading change from the previous timestep. Its reward function considers the target point, running time and turning angle at the same time. Compared with the LSTM structure, the Gated Recurrent Unit (GRU) network has fewer parameters, which helps to save training time. A series of simulation results show that the proposed deep deterministic policy gradient (DDPG) algorithm can obtain excellent results in simple and complex environments.

Pricing Analysis for Railway Multi-Ride Tickets: An Optimization Approach for Uncertain Demand within an Agreed Time Limit

A multi-ride ticket with a certain period of validity and maximum number of uses has been introduced into railway transport. The key to pricing the railway multi-ride ticket is determining the uncertain demand within an agreed time limit. Unfortunately, limited studies have focused on this pricing issue. Therefore, we focused on railway multi-ride ticket pricing optimization in two different scenarios: a single train with multiple stops and multiple trains with multiple stops. First, the expected coefficient and incentive coefficient were introduced to describe the decision-making process for multi-ride tickets and simulate the change in passengers’ travel behavior after purchasing multi-ride tickets. Then, passenger demand functions based on a normal distribution were developed to establish the pricing models with maximized revenue. Finally, we adopted improved particle swarm optimization (PSO) to solve the models. Two numerical cases were used to verify the models separately for two application scenarios. The results revealed that the multi-ride ticket pricing problem is not a simple summation of pricing for one-time travel of passengers. In the situation of a single train with multiple stops, the expected coefficient is positively related to the total income, whereas the incentive coefficient has limited influence on the optimal price and total revenue. Furthermore, a multi-ride ticket should allow the passenger to take trains eight times at most in 8 days at the price of CNY 4922 (abbreviated as 4922 (8, 8)) rather than 3785 (8, 6). Railway enterprises should cautiously limit the scope of trains available for multi-ride tickets in the case of multiple trains with multiple stops.

3D simulation model for IoD-to-vehicles communication in IoD-assisted VANET

Vehicle ad hoc networks (VANETs) have gradually emerged to enhance transportation information, entertainment, safety, and other services. However, such infrastructures have certain limitations, causing intermittent network disconnection. Further, in urban areas, terrain heights act as obstacles and hinder or attenuate transmitted signals. In this study, we propose a dynamic 3D internet of drones collaborative communication approach for efficient VANET-assistance (3DIoDAV) by integrating the IoD network and VANET to support terrestrial communication. We model IoD locations as an optimization problem to optimize the IoD nodes in three-dimensional terrain. Improved particle swarm optimization is used to optimally deploy IoD nodes in 3D terrain for minimizing the number of isolated vehicles. The proposed approach considers the terrain profile influence on communication. Therefore, we propose a 3D propagation model for efficient IoD-to-vehicle (IoD2V) communication in 3D space. Experiments are performed based on the received signal from ground vehicles to examine the performance of the proposed model and the 3DIoDAV approach. Simulation results show different behaviors of IoD nodes in two-dimensional (2D) and 3D scenarios. Comparison with 2D VANET-assisted and IoDAV approaches demonstrates the proposed 3DIoDAV approach’s ability to detect terrain obstacles, which guarantees the dispatching of IoD nodes into the most appropriate locations in 3D space, thereby minimizing the impact of terrain obstacles on communication.

Improved particle swarm optimisation tuned PID position control of voice coil semi-active electrohydraulic damper

Control on the variable damping coefficients in the semi-active damper enhances ride comfort and vehicle stability. The semi-active electrohydraulic damper (SEH) presented in this study incorporates voice coil-based robust valve control to modify the damping coefficients. However, the position tracking accuracy of the voice coil actuator (VCA) gets negatively influenced by the nonlinearities such as load variations due to external disturbances, back emf, and dynamic damper characteristics. Therefore, to achieve excellent position control, a restart particle swarm optimisation with the Gompertz function (PSO-RG) is presented to obtain optimal tuning control gain for the proportional integral derivative controller. The PSO-RG algorithm applies Gompertz function-based particle coefficients and restarting technique. These techniques allow the PSO-RG to overcome the limitations of standard PSO, such as premature convergence and particle entrapment in local minima. The algorithm is tested in five selected benchmark functions for 10, 30 and 50 dimensions. The test data conveys that PSO-RG outperforms the standard PSO in mean, standard deviation, the best solution, and convergence speed. The transient response on identified and validated model of VCA demonstrated a 75.21% reduction in integral absolute error value than conventional PSO. The obtained maximum sensitivity and total variation values proved the robustness and smoothness of PSO-RG tuned PID control. Furthermore, the experimental analysis on the quarter car model shows that PSO-RG tuned PID precisely tracked the reference position with a 21.66% decrease in root mean square error to conventional PSO tuned PID. The CarSim co-simulation test results in the large van model indicate the need for precise VCA position control in enhancing ride comfort and improving the roll stability of the vehicle.

Building Carbon Emissions Prediction Based on Deep Learning Network with Improved Particle Swarm Optimization

Building Carbon Emissions Prediction Based on Deep Learning Network with Improved Particle Swarm Optimization

Research on the control strategy of DC microgrids with distributed energy storage

As a supplement to large power grids, DC microgrids with new energy access are increasingly widely used. However, with the increasing proportion of new energy in DC microgrids, its output fluctuations directly affect the overall stability of the microgrids. Distributed energy storage can smooth the output fluctuation of distributed new energy. In this paper, an AC-DC hybrid micro-grid operation topology with distributed new energy and distributed energy storage system access is designed, and on this basis, a coordinated control strategy of a micro-grid system based on distributed energy storage is proposed. To maintain the voltage stability of the DC bus and make each station have the power-sharing ability, the AC/DC flexibly interconnected converter should adopt two control strategies. The power can flow bidirectional in the power scheduling and distribution of the energy storage station; At the same time, different power distribution schemes will generate different scheduling costs. To optimize the operation of energy storage power stations, an improved particle swarm optimization algorithm is adopted in this paper to optimize the scheduling task allocation scheme. The optimization objective is the lowest scheduling cost, to realize the optimal scheduling of energy storage power stations. In this paper, based on a Matlab/Simulink environment, a microgrid system based on an AC-DC hybrid bus is built. The simulation results verify the effectiveness of the proposed microgrid coordinated control strategy.

Decoupling control of bearingless brushless DC motor using particle swarm optimized neural network inverse system

Neural network inverse system decoupling control can effectively solve the nonlinear strong coupling problem of bearingless brushless DC motor, but it has the problem of slow convergence and easy to fall into local extreme value. An Improved Particle Swarm Optimization (IPSO) algorithm has been employed to optimize the initial weights of the BP neural network inverse system. Comparisons between traditional inverse system decoupling control and the proposed method through simulations were conducted to confirm the efficacy and superiority of the proposed decoupling strategy. Furthermore, experiment research indicates that effective decoupling control can be achieved for both speed and radial displacement, as well as between the x and y-axis radial displacements, showcasing the good dynamic decoupling performance and stability of the proposed decoupling control strategy.

Analysis and prediction of surface topography characteristics and influence factors of tool passive vibration in milling process

The surface topography of the processed workpiece has a significant impact on its service performance, and the tool undergoes passive vibration due to the influence of milling forces during the machining process. This article focuses on the influence of milling parameters and tool passive vibration on the formation process of surface topography. Firstly, the forming mechanism of surface topography during passive vibration of cutting tools was studied, and a cutting edge motion trajectory model considering milling parameters and passive vibration of cutting tools was established; And the influence of milling parameters on surface topography with and without tool passive vibration was analyzed through experiments and simulations; A prediction model for the maximum height S z and areal arithmetic mean height S a of surface topography was established using least squares support vector machine (LSSVM). We used the Improved Particle Swarm Optimization (PSO) algorithm to search for optimal solutions for kernel width coefficients and regularization parameters in LSSVM, and wrote a program to improve the PSO-LSSVM prediction model. The results indicate that the proposed prediction model can provide a certain basis for the selection of actual milling experimental parameters.