Particle Swarm Optimization Procedure Research Articles

Online identification of Hammerstein systems with B‐spline networks

SummaryNonlinear systems widely exist in real‐word applications and the research for these systems has enjoyed a long and fruitful history, including the system identification community. However, the modeling for nonlinear systems is often quite challenging and still remains many unresolved questions. This article considers the online identification issue of Hammerstein systems, whose nonlinear static function is modeled by a B‐spline network. First, the identification model of the studied system is constructed using the bilinear parameter decomposition model. Second, the online recursive algorithms are proposed to find the estimates using the moving data window and the particle swarm optimization procedure, and show that these estimates converge to their true values with a low computational burden. Numerical examples are also given to test the effectiveness of the proposed algorithms.

Techno‐economic optimization framework of renewable hybrid photovoltaic/wind turbine/fuel cell energy system using artificial rabbits algorithm

AbstractIn order to maximize the electricity supply from clean energy sources, the goal of the smart power system is to unite all renewable energy sources. The goal of the present study is to use three optimization techniques, artificial rabbits optimization algorithm (ARO), grey wolf optimizer (GWO), and whale optimization algorithm (WOA), to reduce the cost of electricity (COE) while improving the reliability of the power supply for rural areas. While using the same control variables for the optimization methods and load profile, various hybrid system configurations are explored. Photovoltaic, wind turbine, fuel cell, and electrolyser systems are all involved in the proposed hybrid renewable system. The ARO methodology is more effective than the GWO, WOA, and PSO procedures in terms of net present cost (NPC) and cost of energy (COE) generation, according to data comparing the three optimization techniques with the traditional Particle Swarm Optimization (PSO) method. The proposed ARO reached a value of COE of 0.4412$/kWh compared to 0.4438$/kWh for GWO, 0.4443$/kWh for WOA, and 0.44378$/kWh for PSO.

A Real-Time Tourism Route Recommendation System Based on Multitime Scale Constraints

In order to increase the capability of real-time intelligent recommendation of tourists’ information on cross-regional city-level tourist routes with epidemic normalization, a real-time intelligent recommendation algorithm for cross-regional city-level tourist routes with epidemic normalization based on multi-time scale constraints is proposed. Under the training of limited samples, the tourist correlation model of the epidemic normalization of cross-regional city-level tourist routes is created. In addition, two kernel functions i.e. the mixed and the global are assembled to excerpt the correspondence features of the epidemic normalization cross-regional city-level tourist route recommendation information. As a result, the well-known particle swarm optimization (PSO) procedure and algorithm with multitime scale constraints are adopted to carry out the adaptive learning of the epidemic normalization cross-regional city-level tourist route recommendation, and the convergence control of the recommended method is comprehended through mining the geographic information data sets of cities. This paper analyzes the universality and ergodicity of tourists' personal interest preferences and social characteristics in urban tourism and combines a gradient algorithm to carry out particle swarm evolution and self-adaptive optimization for the recommendation of cross-regional city-level tourist routes with a normalized epidemic situation, so as to realize the group real-time intelligent recommendation of tourists’ information on cross-regional city-level tourist routes with the normalized epidemic situation. The model outcomes indicate that the exactitude and precision of cross-regional city-level tourism route information recommendation with this algorithm are decent, and the convergence of the swarm intelligence optimization (SIO) problem is robust, which can circumvent dipping into the local optimal solution in the process of real-time intelligent recommendation of tourism routes and improve the intelligence and global stability of cross-regional city-level tourism route recommendation with epidemic normalization.

Development of a New Backward Directional Coupler Based on Perforated Substrates

In this paper, two wideband 10 dB backward directional couplers based on artificial perforated substrates over the frequency range of 25-35 GHz and 32-38 GHz are developed. An analytical method is proposed to design the coupler geometrical parameters. The theoretical modeling is established based on the coupled version of the transmission line (TL) theory using the extended version of the ABCD matrix for four ports microwave network. It is shown that using the proposed method, all required parameters of the directional coupler are determined using the per-unit-length of the applied lines. The geometrical parameters of primary designed couplers are optimized using the particle swarm optimization (PSO) procedure to improve the performance of couplers. The designed couplers are also simulated using High Frequency Structure Simulator (HFSS) software. Moreover, sensitivity analysis is carried out to investigate the effect of fabrication imperfections of the proposed couplers. The obtained results show that the simulated results agree well with the theoretical ones and a low insertion loss (IL) with high return loss is obtained over a wide frequency range bandwidth.

Automating reference temperature measurements for crop water stress index calculations: A case study on grapevines

The use of temperature as an indicator of water stress is gaining attention in agriculture. However, the need for reference temperature measurements in the field (Twet and Tdry) for normalization purposes (calculation of the crop water stress index - CWSI) inhibits the practical implementation of this technique. Therefore, in this study, two new models namely the heat transfer (HT) model and the empirical (EMP) model, are presented and compared with the standard method of physical reference temperature measurements and leaf energy balance calculation as an exploratory case study on grapevines. The HT model is a novel method, based on physical heat transfer principles and uses only input data obtained from a conventional weather station to determine reference temperatures. The EMP model is a simple method based on wet- and dry-bulb temperatures, calculated from ambient temperature and relative humidity measurements. To develop and evaluate the new models, physical measurements of the reference temperatures were taken in a commercial vineyard cv. Cabernet Sauvignon under different levels of water stress at two times during the day on seven days over the growing season. These physical measurements were used to optimize unknown parameters in the new methods by using the particle swarm optimization procedure. Input data for the model was collected by a conventional weather station located nearby the experimental vineyard block. In the validation process, it was found that the HT model can accurately predict the reference temperatures to within 0.5 °C and 1.0 °C for Twet and Tdry, respectively, reacting to environmental conditions as expected. The EMP model, requiring even less meteorological information, can accurately predict the reference temperatures to within 0.8 °C and 1.4 °C for Twet and Tdry, respectively. These proposed methods can provide reference temperatures that do not require physical measurements which can make the use of CWSI more practical and easier to implement for determining plant water stress in vineyards.

Trading cryptocurrencies using algorithmic average true range systems

AbstractThis research makes the first attempt to design, optimize and use average true range (ATR)‐based trading systems for five popular cryptocurrencies. We used particle swarm optimization procedures to optimize systems with multiple objectives that are based on the ATR concept. Our aim was to determine the best configurations for each system that would maximize net profits, the profit factor, and the percentage of profitable trades. We demonstrate that the ATR‐based systems can predict the price trends of the examined cryptocurrencies. Our results also indicate that optimized Keltner Channel‐based systems improve the ability of the stand‐alone optimized ATR systems to forecast trends, net profits, and the profit factor. Finally, both systems perform better for long trades than for short trades.

Surface Roughness Prediction and Optimisation using Novel Joint Artificial Neural Network and Bat Algorithm

This paper targets the surface roughness concept in end milling in which the tool-work material combination is central to its success. At present, sufficientoptimal surface roughness information is repeatedly not accessible to CNC end milling operators and this problem is anticipated to grow worse in the forthcomingyears. Consequently, the unique development and validation of optimisation tools are interventions to tackle access to optimal roughness information problems. This paper examined two novel models, the combined artificial neural network and bat algorithm as well as joint artificial neural network and particle swarm optimisation to predict and optimise the process parameters of an end milling scheme. Both models were tested with literature data. Additionally, the work investigates machining time and introduces a bi-objective fuzzy goal programming optimisation model. The striking results revealed the optimal values as 0.8816 and 0.8088 for the particle swarm optimisation procedurewhile the bat procedureyielded 0.275 and 0.178, which places the bat procedureahead of the counterpart, particle swarm optimization procedure.

Application of the PSO for the construction of a 3-axis stable magnetically actuated satellite angular motion

Particle swarm optimization (PSO) is a non-gradient biologically inspired global optimization method. It utilizes direct calculation of the optimized cost function during the numerical simulation of the process in question. Significantly complex and implicit functions that prohibit clear gradient expression are effectively handled by PSO. This feature is utilized in the paper to construct specific angular trajectory of a magnetically actuated satellite. A cost function is suggested that confines the control torque corresponding to this trajectory to the plane orthogonal to the geomagnetic induction vector. Optimal trajectory coefficients are found by PSO. The control structure is based on the Lyapunov feedback approach and explicitly accommodates the gravitational and aerodynamic torques influence. PSO procedure provides the optimal control gains on this stage. Stability of the resulting motion is demonstrated. Numerical simulation example with different disturbance factors illustrates the expected stabilization accuracy of the order of few degrees.

Smart fault detection and optimal variables identification using Kernel Mahalanobis Distance for industrial manufacturing processes

ABSTRACT The development of a new methodology for the accurate fault detection considering only those really important features of practical impact is the main motivation of this work. Therefore, the Kernel-based Mahalanobis-Taguchi System as a pre-classification stage followed by a binary particle swarm optimization procedure is introduced in this work in order to identify variables of impact by means of a feature screening in an optimal manner for industrial manufacturing processes. In order to study the practical efficiency of the presented approach, a computational comparison between the introduced approach and the standard Mahalanobis-Taguchi System is reported. Furthermore to test the relevance of the selected features by the proposed fault diagnosis approach, the sequential feature selection (SFS) and Random Forest (RF) algorithms are implemented and five experiments were carried out training a total of 15 machine learning algorithms. The presented approach is applied to foam injection and welding processes from the manufacturing automotive industry showing that the presented approach can achieve betters accuracy values for fault detection by selecting a small subset of features than the other presented feature selection techniques.

AN OPTIMIZED ADAPTIVE OVERCURRENT RELAY PROTECTION IN DISTRIBUTION SYSTEM USING CHAOS BASED SATIN BOWERBIRD OPTIMIZATION ALGORITHM (C-SBO)

The increased availability of protections of distributed system is a major role by ensuring the continuity of electrical power supply. The optimization process for a distributed system is to coordinate with overcurrent relay protection with adaptive overcurrent relay model. Here the protection of the system is an important factor since the protection methods play a vital role in distributed systems. The conventional methods discussed the current in the power system to reduce the fault current tolerance value by using various optimization algorithms like Antlion and Butterfly optimization techniques. The distributed three-phase system uses the IEEE bus network with adaptive overcurrent relay model, which produces the best output as compared to the conventional method. Here the particle swarm optimization (C-SBO) procedure is used to improve the relay setting and it reduces the fault current across the relay.

Swarming morlet wavelet neural network procedures for the mathematical robot system

The task of this work is to present the solutions of the mathematical robot system (MRS) to examine the positive coronavirus cases through the artificial intelligence (AI) based Morlet wavelet neural network (MWNN). The MRS is divided into two classes, infected I(θ) and Robots R(θ). The design of the fitness function is presented by using the differential MRS and then optimized by the hybrid of the global swarming computational particle swarm optimization (PSO) and local active set procedure (ASP). For the exactness of the AI based MWNN-PSOIPS, the comparison of the results is presented by using the proposed and reference solutions. The reliability of the MWNN-PSOASP is authenticated by extending the data into 20 trials to check the performance of the scheme by using the statistical operators with 10 hidden numbers of neurons to solve the MRS.

A case study for the optimization of moment-matching in wind turbine blade fatigue tests with a resonant type exciting approach

In order to effectively test the performance of the wind turbine during the blade fatigue test with a resonant type of exciting approach, it is important to moment-match the test moment to the target moment. The distribution of the moments on the critical sections can be adjusted by using additional masses that are reasonably located on these sections, but that have different location, number and size options. Achieving this through the traditional trial-and-error method however, which attempts to find the best additional mass variables combination for optimal moment-matching, requires an extremely heavy artificial workload, as the combination options for the additional mass variables are many. To tackle this dilemma, FEA (Finite Element Analysis) associated with PSO (Particle Swarm Optimization) method is now commonly used for moment prediction and optimal combination searching. Nonetheless, the enlargement of the blade scale has meant that using FEA poses some disadvantages, which in turn means an increased amount of time needed for integrated-parameters computing. Moreover, pre-computing work, such as physical object modeling, grid generation & checking, solution setting & parameter adjusting, may also require lots of time and a heavy reliance on professional workers to carry out processing tasks – requirements that are adverse to realizing an automatic dealing procedure. Fortunately, in the blade design and factory leaving stages, those important blade parameters are already defined in a lumped manner. This therefore leads to an applicative use of TMM (Transfer Matrix Method) as when outputting the section data, including moment, it only requires these section parameters for the modeling and the moment-matching process. It is reasonable to believe that FEA usually offers better computing accuracy, while it is also understandable that for the present condition cloud and after describing the lumped definition manner, TMM also presents sufficient accuracy. For the present case study, TMM shows satisfactory prediction accuracy when compared with the test results, with a relative error of no more than 1% for any critical blade section.Additionally, moment-matching optimization work has been conducted by combining with the PSO procedure. The optimal variables combination is finally confirmed with two additional masses of 4000 kg and 605.6 kg located 25m and 48m from the blade root section respectively, which obtains the minimum standard of 4.25%, a higher matching precision compared with that of the test which had 5.6%. Thus, it seems that the TMM-associated PSO method introduced in this paper provides a more efficient way to figure out the optimal variables combination for moment-matching while maintaining sufficient computational precision. In addition, since the strict requirements for moment-matching accuracy have not yet been officially issued, the present case study does not treat the additional mass number as a variable but as the constant 2. Compared with the test however, this paper's result still shows improvements in moment-matching accuracy while also reducing the installing workload needed for the 2 additional masses.

WITHDRAWN: A novel methodology to capture the maximum power from variable speed wind turbines using fuzzy proportional integral Controllers, modified PSO-GSA algorithm

This article provides a modified technique for controlling generator torque efficiently in wind turbines (WT). The generator torque is generally used at low wind speeds to obtain extreme power. In this study, we introduce a method for torque of the WT generator is controlled by using a fuzzy proportional integral controller (FPI). To upsurge the PI gain, by using a radial base function neural network (RBF-NN). And another hand we conducted the optimal dataset for training this neural network has been delivered by the techniques of modified procedure of gravity search algorithm and particle swarm optimization (GSA-PSO) algorithm. Thus, the benefits of enhanced technique performance and the reliability of the FPI can be completely utilized in the control course. By showing the simulation of WT experiments with help of Matlab, the performance of a modified GSA-PSO based FPI controller to meet the high accuracy and quick response requirements of permanent magnet direct drive (PMDD) is better than that of a PI controller, Without the use of modified PSO in situations where the wind speed changes abruptly.

Flexible Virtual Cell Design for Ultradense Networks: A Machine Learning Approach

With the ever-growing demand for even higher throughput, ultradense networks (UDNs) are being deployed for the fifth generation (5G) mobile communications. Although massively distributed radio access points (APs) result in a considerable increase in throughput, they also cause some critical problems. When employing a wireless backhaul, the backhaul capacity becomes a limiting factor, which may result in a high packet loss rate. Furthermore, dense deployment of APs leads to more frequent handoffs for mobile user equipments, which results in heavy measurement and signaling overhead. To address the problem of frequent handoffs, virtual cell (VC) has been considered as a promising solution. However, the limited wireless backhaul capacity encountered by inflexible VC design may still result in an intolerable packet loss rate. For a better trade-off between the packet loss rate and the handoff overhead, a machine learning approach for flexible VC design is proposed that leverages particle swarm optimization (PSO) to quickly find the optimal VC solution. To be responsive to the dynamic traffic demand and backhaul capacity of APs, a new parameter called “weighted distance” is employed in the modified K-means algorithm, which is nested in the PSO procedures for master AP selection and VC boundary determination. Compared with an exhaustive search, optimal VC solutions can be found efficiently through considerably fewer iterations. The proposed method is generic and applicable to disparate UDN application scenarios.

Particle Swarm Optimization and L´evy Flight integration

Particle Swarm Optimization (PSO) is a well-known and popular stochastic optimization method. The Lévy Flight (LF) properties were used to improve the canonical PSO known as premature convergence. The Lévy Flight was applied to change each particle walk on the fitness landscape. We analyze the literature modifications that concluded that Levy flight improved the PSO providing better search space exploration. Based on this conclusion, we propose new approaches to integrate Lévy Flight with PSO by changing initial points in the search space and learning strategies as inertia and constriction coefficients. We use seven standard test functions for an experimental evaluation and scores based on ranking to compare PSO variants. The ranked benchmarks were average performance, standard deviation, and best and worst found solutions obtained from multiple trials. The main contributions are a systematic overview of LF modifications applied in PSO and three new LF applications in canonical PSO procedure. The new approaches are swarm initialization based on LF, lower dimension LF inertia coefficient, and LF-based constriction factor. Another contribution is numerical evaluations on various benchmark functions with diverse characteristics. Two of the proposed modifications performed better or equal, and the third was only 2% worse than the best canonical PSO from the trial.

Simulations of radiation effects on erbium–ytterbium co-doped fiber amplifiers for space applications

An improved model of erbium–ytterbium codoped fiber amplifier (EYDFA) considering the effects of radiation is proposed. In this model, terms of radiation-induced losses are introduced into power propagation equations, combined with the simplified rate equations to simulate radiation effects on EYDFA performances related to space missions. A method for recovering background losses and radiation-induced losses of EYDFA is illustrated. Numerical simulation tools are based on a homemade particle swarm optimization procedure. In the experimental aspect, two amplifiers of the same type were irradiated with Co60 gamma rays up to a total dose of 10 and 50 krad, respectively. Compared with the experimental data, the curve of output power drawn by the model is in good agreement with the experimentally measured values, validating the validity of the theoretical model. Using this method that combined experimental testing of online monitoring with theoretical simulation provides a new way to evaluate the performance degradation of commercial amplifiers in a radiation environment.

ОПТИМІЗАЦІЯ РОЗТАШУВАННЯ ДЕМПФЕРІВ СУХОГО ТЕРТЯ ДЛЯ СЕЙСМІЧНОГО ЗАХИСТУ КАРКАСНИХ БУДІВЕЛЬ ЗА ДОПОМОГОЮ МЕТОДУ РОЮ ЧАСТИНОК

Abstract. A challenging problem of Civil Engineering is the protection of buildings against dynamic loads and earthquake impacts. The advanced solutions employ lightweight structures equipped with special damping devices. These devices can be active or passive and their application depends on the investments for the project. The active devices change their properties depending on the structural response and they are the most expensive ones. On the other hand, passive devices are essentially cheaper and, in many cases, require minimal costs of installation and maintenance. Last decades, passive friction dampers are widely used for the earthquake protection of multi-storey buildings. The friction dampers make use of the effect of solid friction to dissipate the mechanical energy and to reduce the amplitude of the vibration of the structure. The friction is developed between two solid bodies sliding in relation to one another. As usual, pairs of metal, polymer or concrete components can be utilized. Determination of the optimal location of the friction dampers inside the building presents a complicated task for the practical design. In this paper, a new approach is developed basing on the method of particle swarm optimization (PSO). The PSO method presents an artificial simulation of the phenomenon of collective intelligence, which is observed in many decentralized biological systems like ant colonies, bee swarms, flocks of birds and even social groups of human individuals. As an illustrative example, the 2D model of a ten-storey concrete frame building is considered. The purpose of the analysis is to minimize the objective function, which is the amplitude of the displacements of the top of the structure in a case of the resonance dynamic load with a frequency of the first normal mode. Non-stationary dynamic analysis is performed by the finite elements method using the program package «LIRA-SAPR» and its module «Dynamic-plus». The software implementation of the PSO procedure is developed using the open-source computer algebra system «Maxima». The optimal placement of friction dampers is determined providing the minimal displacements of the frame. The developed approach can be further extended to various problems of the optimal design of buildings and structures.

A Hybrid Approach for the Synthesis of Nonuniformly-Spaced Linear Subarrays

By combining convex programming (CP) and particle swarm optimization (PSO), a hybrid method for the synthesis of excitations and locations of nonuniformly spaced linear subarrays to minimize the subarray number is proposed and discussed in this article. The synthesis problem herein is formulated as a CP problem by minimizing the l <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> -norm with respect to the excitation variables, and the PSO procedure is carried out as far as location variables are concerned. Through collaborations like this, we can finally get the optimal solution in a global sense. A set of representative numerical experiments shows the effectiveness of this method with quite a number of subarrays saved when compared with the uniformly spaced layouts.

A systematic sampling evolutionary (SSE) method for stochastic bilevel programming problems

Stochastic bilevel programming is a bilevel program having some form of randomness in the problem definition. The main objective is to optimize the leader’s (upper level) stochastic programming problem, where the follower’s problem is assumed to be satisfied as part of the constraints. Due to the involvement of randomness property and the hierarchical nature of the optimization procedure, the problem is computationally expensive and challenging. In this paper, a new meta-heuristic type algorithm is proposed that can effectively solve stochastic bilevel programs. The algorithm is based on realizing the random space, systematic sampling technique to choose a representative action from the leader’s decision space and on a hybrid particle swarm optimization procedure for searching its corresponding follower’s reaction for each leader’s action until Stackelberg equilibrium is achieved. The algorithm is shown to be convergent and its performance is checked using test problems from literature. The simulation result of the algorithm is very much promising and can be used to solve complex stochastic bilevel programming problems.

A compact sharp-edge wide bandpass near-infrared filter based on graded-index sinusoidally corrugated photonic crystal

In this study, a wide-bandpass near-infrared filter is devised based on a novel sinusoidally corrugated multilayer structure. The presented structure exhibits a wider bandpass and higher average transmission amplitude than the conventional triangular design. Cut-on frequency of the proposed bandpass filter can be tuned by the variation of horizontal lattice constant. Thus it can be applied as a tunable multichannel filter. Particle swarm optimization procedure was carried out to find the optimal arrangement of dielectric materials to eliminate the undesired fluctuations in the transmission spectrum at the beginning of the passband region. We compared the bandpass filtering behavior of the simple/graded sinusoidal photonic structures with and without anti-reflective double layer coating. Utilization of the optimized graded-index Quasi-periodic photonic crystal results in a sharper cut-on wavelength as well as flatter transmission spectrum than the application of the anti-reflective double-layer coating.