Unified Particle Swarm Optimization Research Articles

Day-ahead photovoltaic power prediction based on a hybrid gradient descent and metaheuristic optimizer

The integration of Photovoltaic (PV) power into energy market requires the development of accurate PV power forecasting models. Apart from the prediction model the selection of the proper optimizer may contribute towards the forecasting accuracy increment. At this paper, we propose a hybrid Adam- Unified Particle Swarm Optimization (UPSO) optimizer to increase the accuracy of day-ahead PV power forecasting. At first, stage the gradient descent based Adam optimizer is executed to initialize the position of the particles. Then, the metaheuristic UPSO is executed for the model’s training. The proposed optimizer is employed to a Long Short-Term Memory model and its efficiency is compared to stand-alone Adam and UPSO. To confirm the robustness of the proposed optimizer, Adagrad, Adadelta, Genetic Algorithm and Artificial Bee Colony optimizers have been also examined. The proposed optimizer is tested on two datasets and four test cases. The results highlight that Adam-UPSO outperforms for 50% of the examined test cases. It may reduce the forecasting error up to 15.21% compared to Adam optimizer and up to 22.6% compared to the persistence model.

Optimal scheduling of prosumer's battery storage and flexible loads for distribution network support

AbstractHere, an efficient Energy Management System (EMS) for residential installations is proposed. The EMS handles on site Photovoltaic (PV) power production along with Battery Energy Storage System (BESS) and performs load shifting under a Demand Side Management (DSM) scheme that considers user comfort. The goal of the proposed methodology is the minimization of power exchange between the prosumer and the grid. In this way, the power flow across the Distribution Network (DN) is decreased, thus reducing power losses, and improving the voltage profile along the DN. Moreover, the EMS increases the self‐sufficiency of the residence. The minimization problem is solved via a metaheuristic algorithm, namely Unified Particle Swarm Optimization (UPSO) utilizing real consumption patterns for residential appliances. The latter adds complexity to the problem but offers accurate modelling of the power demand and improves the performance of the EMS scheme. The proposed EMS is locally implemented as a plug‐and‐play scheme and benefits both the prosumer and the DN under low computational burden, proving to be efficient. Simulation results for a case study verify the efficiency of the proposed EMS and the comparison with other similar methods proves its superior performance.

UPSO-FSVRNET: Fuzzy Identification Approach in a VANET Environment Based on Fuzzy Support Vector Regression and Unified Particle Swarm Optimization

UPSO-FSVRNET: Fuzzy Identification Approach in a VANET Environment Based on Fuzzy Support Vector Regression and Unified Particle Swarm Optimization

Multi-swarm UPSO algorithm based on seed strategy for atomic clusters structure optimization

Particle Swarm Optimization (PSO) algorithm is prone to get trapped in local optima and insufficient information exchange among particles. To solve this problem, this paper proposes a Multi-swarm Unified Particle Swarm Optimization algorithm based on Seed Strategy (SS-DMS-UPSO) to optimize the atomic clusters structure. In this algorithm, the population is divided into some sub-populations evolving randomly and evenly, and each sub-population uses UPSO algorithm with different unification factors to evolve independently in parallel. After a certain number of independent evolution, the particles of all sub-populations are merged into a new population, and the population is again randomly divided into average sub-populations. Iterate the algorithm repeatedly in this way. And finally the global best particle can be obtained. The experimental results show that the SS-DMS-UPSO algorithm can search for the optimal structure or extremely similar optimal structure for atomic clusters with atomic numbers between 2 and 31. For atomic clusters with atomic numbers between 32 and 35, the algorithm can find its approximate optimal structure. Compared with other algorithms, the difference between the lowest energy value and the ideal energy value obtained by the SS-DMS-UPSO algorithm is much smaller. It means that its optimal structure of the atomic clusters is closer to the stable structure, and the algorithm is more stable, which proves the effectiveness of the SS-DMS-UPSO algorithm.

Tailoring Job Shop Scheduling Problem Instances Through Unified Particle Swarm Optimization

Problem instances are paramount when testing the performance of any learning algorithm. For this reason, it is customary to use widespread problems known as benchmark instances. Nonetheless, these are usually generated disregarding heuristics and their nature. For Job Shop Scheduling problem, researchers have created such instances based on random distributions. This idea may bias conclusions about the algorithm under test since a practitioner can only observe performance from a limited perspective, which may not even reflect real-life situations. However, addressing this issue implies tackling the instance generation problem while considering the nature of the solution approach. Hence, in this work, we propose an instance generator based on the Unified Particle Swarm Optimization algorithm, which can tailor instances to different goals. To validate our approach, we include instances generated to different heuristics and instances tailored to a variety of features. In the first case, we seek to favor or hinder one heuristic whereas doing the opposite for the remaining ones. In the second one, we explore instances with specific feature values. Our data reveal that the proposed approach fulfills the expectations and can effectively deal with different kinds of instances. We analyse the nature of the generated instances and their insights, which can be used to further the study about heuristics and problem features.

Vibration-Based Delamination Detection in Composite Structures Employing Mixed Unified Particle Swarm Optimization

The present paper introduces mixed unified particle swarm optimization (MUPSO) algorithm to detect and quantify the delamination damages in composite beams and plates using vibration responses. MUPSO is a combination of conventional continuous unified particle swarm optimization (UPSO) and a binary version of UPSO, which is proposed in this paper. The first few modes have been used to detect delamination damages through a two-stage method. In the first stage, rough locations of damages have been determined from a mode-shape curvature-based damage index using the gapped smoothing method. In the second stage, the mixed UPSO approach is used to get an accurate lengthwise location of the delamination along with the delamination interface. While the continuous part of the mixed UPSO has been used to obtain the lengthwise location of the damages, the binary part of the mixed UPSO has been used to determine the interface of the damage. The present method has quantified each damage scenario consisting of single and multiple interface delaminations with acceptable accuracy. The algorithm is also found to be performed satisfactorily under a noisy environment. The results obtained from experimental investigation are also presented. Overall, the present two-stage method has shown very good potential in the field of damage detection of composite structures.

Particle Swarm Optimization for Computing Nash and Stackelberg Equilibria in Energy Markets

Interactions among stakeholders in deregulated markets lead to complex interdependent optimization problems. The present study is motivated by load control programs in energy markets and more precisely by using the power supply interruption as a tool for reducing consumers’ demand voluntarily, also known as voluntary load curtailment programs. The problem is formulated as a Stackelberg game, specifically, as a bilevel optimization problem that belongs to the mathematical programs with equilibrium constraints. In this game, a player that acts as leader determines the actions of the players that act as followers and play a Nash game among them through a subsidy program. The corresponding equilibria need to be found and the presence of nonconvex functions makes the use of metaheuristic algorithms attractive. An extension of particle swarm optimization is proposed for solving such problems based on the unified particle swarm optimization that is a variation of the plain particle swarm optimization algorithm. The proposed algorithm is tested by solving some examples of the formulated games in order to study its efficiency and the interactions between the stakeholders of the market.

Assessment of Optimum Location of Non-circular Failure Surface in Soil Slope Using Unified Particle Swarm Optimization

The solution of slope stability problems requires determination of critical failure surface and associated optimum/minimum factor of safety (FOS) value. Investigation of critical failure surface in any soil slope can be posed as a constrained global optimization problem. In this article, a simplified technique of generation of segmented non-circular failure surface is proposed. Unified particle swarm optimization (UPSO) method is utilized to search for the critical failure surface and associated minimum FOS of the slope. Limit equilibrium technique-based Morgenstern–Price method is used to evaluate the FOS value of the potential failure mass which satisfy both force as well as moment equilibrium. To demonstrate the performance of UPSO and its competence in yielding critical failure surface along with minimum FOS value, three benchmark problems with differing complexities are chosen from existing literatures. The results obtained are compared and are found to be in accord with the published literatures. The performance of UPSO as a solution algorithm for slope problems are established through convergence studies of various related parameters.

Cost optimization of rectangular RC footing using GA and UPSO

The paper presents an estimation of the optimum cost of an isolated foundation following the safety and serviceability guidelines of Indian Standard (IS) 456:2000. Two adaptable optimization algorithms are developed for the first time to optimize the cost of any type of isolated footing design. Two optimization methods, i.e., constrained binary-coded genetic algorithm, with static penalty function approach and unified particle swarm optimization are developed in MATLAB compliant for optimal design of any isolated foundations. The objective function formulated is based on the total cost of footing. This includes the cost of concrete, the cost of steel and cost of formwork. The design variables which influence the total cost of footing are plan area and depth of footing and area of flexural reinforcement at moment critical sections. The footing design algorithm is developed according to the biaxial-isolated rectangular footing as per IS codes. The constraints, e.g., dimension of footing, restriction on bending, shear stresses and displacements, are considered in the footing design algorithm which acted as a subroutine to the developed optimization programs. Four different numerical examples have been solved to evaluate the versatility of the developed method. A comparison study has been done to observe the efficacy of both the optimization methods.

Evaluation of PNN pattern-layer activation function approximations in different training setups

The processing of inputs in the first two layers of the probabilistic neural network (PNN) is highly parallel which makes it quite appropriate for hardware implementations with FPGA. One of the main inconveniences however remains the implementation of the nonlinear activation function of the pattern layer neurons. In the present study, we investigate the applicability of three approximations of the exponential activation function with look-up tables of different precision and the effect this has on the training process and the classification accuracy. Furthermore, seeking for a highly-parallel hardware-friendly algorithm for the automated adjustment of the spread factor $$\sigma _i$$, we investigated the performance of fifteen PNN training setups, which are based on the differential evolution (DE) or unified particle swarm optimization (UPSO) methods. The experimental evaluation was performed following a common experimental protocol, which makes use of the Parkinson Speech Dataset, as this research aims to support the development of portable medical devices that are capable to detect episodes with exacerbation in patients with Parkinson’s disease. The performance of the most successful setups is discussed in terms of error rates and from the perspective of the resources required for an FPGA-based implementation.

Frequency-Based Damage Assessment of Composite Members Using Unified Particle Swarm Optimization

A numerical procedure is presented to detect and quantify damages in laminated composite beam and plate-like structures using unified particle swarm optimization (UPSO) technique. The presence of damage in any structure alters its integrity and performance (causes reduction in stiffness and strength) and thus changes the vibration response parameters such as frequency, mode shapes, damping ratio, and frequency response functions (FRFs). These changes in structural response parameters can be used for the identification of flaws in composite structures with the help of an optimization technique. For the present study, natural frequency is used as the diagnostic parameter and UPSO has been used as the optimization technique. An in-house Matlab code has been developed using finite element analysis (FEA) and has been used to extract the vibration responses of undamaged and damaged composite structures. Since composite materials are anisotropic in nature, effect of anisotropic damage on vibration characteristics of damaged composite structures is also presented. It is observed that the anisotropic damage exhibits strong orthogonality and in general, the occurrence of damage changes the vibration characteristics of the structure. It is also observed that as the damage intensity and damage area increase, there is considerable reduction occurring in the natural frequencies. The efficacy of the proposed methodology to identify damages is demonstrated using some numerically simulated composite beam and plate structures with single and multiple damages.

Efficient RES Penetration under Optimal Distributed Generation Placement Approach

In this paper, a novel version of the Optimal Distributed Generation Placement (ODGP) problem regarding the siting and sizing of Renewable Energy Sources (RESs) units is presented, called Optimal RES placement (ORESP). Power losses constitute the objective function to be minimized, subject to operational constraints. The simultaneous installation of a mix of RESs is considered and the Capacity Factor (CF) ratio is used as an aid for taking into account: (a) the geographical characteristics of the area, in which the examined Distribution Network (DN) is placed, (b) the different weather conditions, and (c) the availability of RESs, all of that at the same time, while keeping the problem complexity at minimum. The contribution of this work is that the proposed methodology bypasses the weather uncertainties and, thus, the RESs’ power generation stochasticity and provides an adequate solution with minimum computational burden and time, since the proposed CF use allows solving the problem under a straightforward way. Unified Particle Swarm Optimization (uPSO) is used for solving ODGP and ORESP. Moreover, a sensitivity analysis regarding the CFs variations is performed and finally a comparison of the proposed method with a more realistic one is performed, to consolidate further the claims of this paper. The proposed method is evaluated on RES-region-modified 33- and 118 bus systems.

Unified Particle Swarm Optimization Based Distributed Generation Placement in Radial Power Systems

Unified Particle Swarm Optimization Based Distributed Generation Placement in Radial Power Systems

Passive control of damaged composite laminates with optimized location of piezoelectric fiber composite patches

This work presents a novel technique to enhance the bending, buckling, and post buckling characteristics of a damaged composite plate. This technique uses piezoelectric fiber composite patches (PFCP) for enhancing the performance and thereby reducing the effects of internal flaws. This paper discusses about the employment of PFCPs in their optimized location. Unified particle swarm optimization (UPSO) is employed for optimizing the PFCP locations for maximum critical buckling temperatures. An inverse hyperbolic shear deformation theory is used for advanced accuracy of the analysis of the smart plate. The post buckling behavior is studied with varying the number of optimally located PFCPs using modified Newton Raphson method. The obtained results are matched with those in the existing literature, wherever possible. It is observed that this is a very promising means of designing new smart material applications.

Reconstructing design parameters of a rectangular resonator via peak signal-to-noise ratio and global optimization algorithms

This article proposes an approach for reconstructing physical parameters of a sample in a rectangular resonator during microwave radiation, knowing a priori, its electric field distribution. The inverse problem was solved using two global optimization algorithms and the peak signal-to-noise ratio (PSNR) criterion. First, the Self-regulated Fretwidth Harmony Search algorithm (SFHS) identified suitable resonant frequencies for a given configuration. Next, the unified Particle Swarm Optimization (UPSO) optimized said configuration. Together, they became a maximization strategy of the PSNR through a dual optimization process. Results showed the ability of the approach for estimating the height of each sample block and the resonating frequency of the cavity. This process takes longer to finish as a higher PSNR is demanded (mainly due to the aforementioned dual optimization). Even so, it allows for more similar electric field distributions between both, the direct and inverse problems.

Buckling load enhancement of damaged composite plates under hygrothermal environment using unified particle swarm optimization

Optimization with Unified Particle Swarm Optimization (UPSO) method is performed for the enhancement of buckling load capacity of composite plates having damage under hygrothermal environment which has received little or no attention in the literature. Numerical results are presented for effect of damage in buckling behavior of laminated composite plates using an anisotropic damage model. Optimized critical buckling temperature of laminated plates with internal flaw is computed with the fiber orientation as the design variable by employing a UPSO algorithm and results are compared with undamaged case for various aspect ratios, ply orientations, and boundary conditions. FEM formulation and programming in the MATLAB environment have been performed. The results of this work will assist designers to address some key issues concerning composite structures. It is observed that the degradation of buckling strength of a structural element in hygrothermal environment as a result of internal flaws can be avoided to a large extent if we use these optimized ply orientations at design phase of the composite structure. This specific application proves the contribution of present work to be of realistic nature.

Bending and Buckling Analyses of Composite Laminates with and Without Presence of Damage and its Passive Control with Optimized Piezoelectric Patch Location

This work presents an efficient technique to enhance the bending and buckling characteristics of a smart composite plate. This paper discusses about the employment of piezoelectric fibre composite patches (PFCP) in their optimized location using unified particle swarm optimization (UPSO) for enhancing the performance and thereby reducing the effects of internal flaws. A finite element formulation based on Inverse Hyperbolic Shear Deformation Theory (IHSDT) for handling bending and buckling analysis of a smart composite plate is used in the present work. In addition to the best performance, reduction in weight of piezoelectric material is obtained as we employ a segmented piezo patch to overcome the degradation in buckling strength due to damage in a composite plate, which indeed addresses the design issues.

An optimal high thermal conductive graphite microchannel for electronic device cooling

This article describes the design of an optimal rectangular microchannel made of a high thermal conductive graphite (HTCG). For simulating the proposed microchannel heat sink, the total resistance model and the entropy generation minimization criterion were used. For solving the optimization problem, the unifi ed particle swarm optimization algorithm (UPSO), was used. Results showed a marked effect of using this high thermal conductor when compared to traditional materials, such as aluminum, and while using air and ammonia gas as the working fl uids. It is also reported the relative effect of the constriction, convective and fl uid thermal resistances on the overall equivalent thermal resistance. As a demonstrative example when changing the nature of the coolant, a titanium dioxide nanofl uid was selected. It was found that the Nusselt number is perceptibly lower, when the coolant is a nanofl uid and the material for the making of the microchannel is an HTCG.

P-MEMPSODE: Parallel and irregular memetic global optimization

A parallel memetic global optimization algorithm suitable for shared memory multicore systems is proposed and analyzed. The considered algorithm combines two well-known and widely used population-based stochastic algorithms, namely Particle Swarm Optimization and Differential Evolution, with two efficient and parallelizable local search procedures. The sequential version of the algorithm was first introduced as MEMPSODE (MEMetic Particle Swarm Optimization and Differential Evolution) and published in the CPC program library. We exploit the inherent and highly irregular parallelism of the memetic global optimization algorithm by means of a dynamic and multilevel approach based on the OpenMP tasking model. In our case, tasks correspond to local optimization procedures or simple function evaluations. Parallelization occurs at each iteration step of the memetic algorithm without affecting its searching efficiency. The proposed implementation, for the same random seed, reaches the same solution irrespectively of being executed sequentially or in parallel. Extensive experimental evaluation has been performed in order to illustrate the speedup achieved on a shared-memory multicore server. Program summaryProgram title: p-MEMPSODECatalogue identifier: AEXJ_v1_0Program summary URL:http://cpc.cs.qub.ac.uk/summaries/AEXJ_v1_0.htmlProgram obtainable from: CPC Program Library, Queen’s University, Belfast, N. IrelandLicensing provisions: Standard CPC licence, http://cpc.cs.qub.ac.uk/licence/licence.htmlNo. of lines in distributed program, including test data, etc.: 9950No. of bytes in distributed program, including test data, etc.: 141503Distribution format: tar.gzProgramming language: ANSI C.Computer: Workstation.Operating system: Developed under the Linux operating system using the GNU compilers v.4.4.3 (or higher). Uses the OpenMP API and runtime system.RAM: The code uses O(n×N) internal storage, n being the dimension of the problem and N the maximum population size. The required memory is dynamically allocated.Word size: 64Classification: 4.9.Nature of problem: Numerical global optimization of real valued functions is an indispensable methodology for solving a multitude of problems in science and engineering. Many problems exhibit a number of local and/or global minimizers, expensive function evaluations or require real-time response. In addition, discontinuities of the objective function, non-smooth and deceitful landscapes constitute challenging obstacles for most optimization algorithms.Solution method: We implement a memetic global optimization algorithm that combines stochastic, population-based methods with deterministic local search procedures. More specifically, the Unified Particle Swarm Optimization and the Differential Evolution algorithms are harnessed with the derivative-free Torczon’s Multi-Directional Search and the gradient-based BFGS method. The produced hybrid algorithms possess inherent parallelism that is exploited efficiently by means of the OpenMP tasking model. Given the same random seed, the proposed implementation reaches the same solution irrespective of being executed sequentially or in parallel.Restrictions: The current version of the software uses only double precision arithmetic. An OpenMP-enabled (version 3.0 or higher) compiler is required.Unusual features: The software requires bound constraints on the optimization variables.Running time: The running time depends on the complexity of the objective function (and its derivatives if used) as well as on the number of available cores. Extensive experimental results demonstrate that the speedup closely approximates ideal values.

Optimal rectangular microchannel design, using simulated annealing, unified particle swarm and spiral algorithms, in the presence of spreading resistance

This article focuses on the solution of the mathematical model of a rectangular microchannel through three algorithms: SA (Simulated Annealing), UPSO (Unified Particle Swarm Optimization) and SO (spiral Optimization). The criterion of minimum entropy generation was used for constructing the model, which includes the effect of the thermal spreading resistance. This type of thermal resistance occurs whenever a smaller heat source (e.g. the chip) comes in contact with the base of a larger heat transfer element (e.g. the heat sink). This simulation work was split into two stages. The first one considered different scenarios for a silicon heat sink using air as a working fluid. In this case, volume flow rates between 4.0 and 4.5 × 10−3 m3/s exhibited lower values of entropy generation rate. The second stage considered three materials (silicon, aluminum, and copper) and two working fluids (air and ammonia gas). We found, for this particular set of conditions, that entropy generation becomes minimum, for a given material and working fluid, as long as a volume flow rate of 4.5 × 10−3 m3/s is used. Even so, the best practical configuration found was an aluminum heat sink with ammonia gas. For all the simulations, a marked effect due to the spreading resistance was found, typically ignored in real life applications.