particle swarm optimization algorithm

Due to the increasing use of steel frames in buildings built in Iran and the increase in the price of profiles compared to raw sheets, the use of plate girders in the construction of steel frames has become significantly popular. Connections are one of the basic components in all structures, including steel structures, and are considered the main factor in the integrity of structural systems. The purpose of this research is to determine the optimal dimensions for the end plate connections to reduce the cost using particle swarm optimization algorithm, so that the mechanical limitations related to the bending moment and the resulting initial stiffness, as well as the safety and integrity of the connection, are not compromised. In the optimization model, six design variables including the dimensions and thickness of the end plate, the diameter and location of the bolts were selected to compare the obtained values with the results of the past literature. To calculate the bending moment and rotational stiffness of the connection, the "component method" found in Eurocode 3-part 1-8 was used, and the analysis and optimal design of beam-to-column connections for two-dimensional steel frames was done in the MATLAB computing environment. The output of this program shows completely satisfactory results compared to the results in the past literature, so that the results obtained by using the aforementioned optimization algorithm show the cost of connections A, B, C and D of the investigated frame, respectively cost reductions of 46.24%, 5.29%, 55.10% and 56.44% by compared to the research of Cabrero and Bayo in 2005. Also, the cost of connections A and C of the examined frame decreased by 8.07% and 0.22%, respectively, compared to Diaz et al.'s research in 2012.

This paper investigates the accuracy and convergence rate of different metaheuristic algorithms in determining the stiffness of structural elements using structural modal parameters and defining a suitable objective function. To achieve this purpose, three different structures, including a three-story one-dimensional frame, a six-story one-dimensional frame and a two-dimensional truss, were investigated. The metaheuristic algorithms, employed in this study, were Genetic Algorithm, Particle Swarm Optimization, and Teaching–learning-based Optimization. The objective function utilized in this study consists of two terms; the first part involves the squared difference between the first frequency of the structure obtained from the responses of the investigated structure and the first frequency obtained from the hypothetical stiffness matrix in each generation of algorithms. The second part measures the norm of the difference between the first mode shape of the structure obtained from the responses of the investigated structure and the first mode shape obtained from the hypothetical stiffness matrix in each generation of algorithms. By minimizing the objective function, the Genetic Algorithm, Particle Swarm Optimization, and Teaching–learning-based Optimization determined the element stiffness of the three-story, six-story and truss structures, thus demonstrating the high efficiency of metaheuristic algorithms in resolving unknown parameters of structures. The average run time for the Genetic Algorithm was 3.38 seconds, 4.47 seconds, and 15.73 seconds for the three respective problems. For Particle Swarm Optimization, the times were 3.76 seconds, 6.47 seconds, and 16.76 seconds. The Teaching–learning-based Optimization achieved times of 1.92 seconds, 4.51 seconds, and 12.76 seconds. The Teaching–learning-based Optimization exhibited the highest convergence rate and the lowest error compared to the Genetic Algorithm and Particle Swarm Optimization. For example, in the two-dimensional truss, the values of the objective function in the last iteration of the Genetic Algorithm, Particle Swarm Optimization, and Teaching–learning-based Optimization were , and , respectively. The Particle Swarm Optimization demonstrated an acceptable convergence rate and error compared to the Genetic algorithm. The Genetic Algorithm, however, displayed a significant error rate in determining the stiffness of structural elements compared to the other two algorithms.

The postal network is a network in which the flow of shipments is collected by vehicles from several sources and transferred to hubs; In the last step, the parcels are sent to the final destination by the vehicles. In order to provide this service, the hubs are located to process the parcels and the local routes are planned to pick up and delivery the packages in the hubs. This problem is known as location-routing problem. The current research seeks to locate the hub and the routing of freight carriers on an intercity scale and at the level of Iran. For this purpose, first, the mathematical model of the problem of locating the hubs and routing the cargo carriers is presented, and since the problem of locating-routing is in the category of complex problems with large dimensions, meta-heuristic algorithms have been used to solve it. In previous studies, meta-heuristic algorithms or a combination of them have been used to solve the location-routing problem, but in the study, the location-routing problem has been solved by integrating two genetic algorithms (GA) and the particle swarm optimization algorithm (PSO). The results of this research show that the integration of two algorithms, PSO and GA, provides a better answer than if only one meta-heuristic algorithm is used to solve the problem. It is worth mentioning that in the current research, the data of Tipax Company was used to locate the hubs and local tours routing.

Nowadays, the development of urbanization and the increase of urban population have caused the air to heat up more than in the past and create urban heat islands. Urban heat islands are a phenomenon caused by the urbanization effects, due to which the temperature in the urban environment rises higher than in the suburbs. This phenomenon can cause irreparable damage due to the increasing atmospheric and environmental temperature, such as biological pollution, greenhouse gas emissions, diseases caused by heat, and impact on water quality brought to communities and the environment. This research proposes an effective and efficient approach with the help of remote sensing and optimization algorithms based on replacing the roof covering of an area with less heat-absorbing coverings to reduce the temperature and try to eliminate the heat island phenomenon. In this research, we are trying to reduce the urban heat island effect based on algorithms and statistical parameters affecting the ambient temperature, which has had few studies in past research. Also, using the intelligent optimization method in this field can cause innovation and create better and more accurate results. The new way that this study examines is to change the roof covering of an area with other functional coverings that reduce the air temperature in that area. The coverings that we considered to replace the covering of the roofs to moderate and cool the temperature of the studied area are two types of coverings: soil and vegetation.

The proposed approach of this research is to use two optimization algorithms of genetic and particle swarm, and the parameters that form the objective function of these two algorithms are the temperature standard deviation and the average financial cost of the coverage changing of each building parcel. The research dataset is Landsat 8 satellite images of Andisheh neighborhood in Tehran. This research uses satellite images for purposes such as preparing color images, mapping the vegetation and non-vegetation indices of the study area, and calculating the earth's surface temperature and urban heat islands.

The results indicate that both optimization algorithms have provided good performance and improved the problem parameters, but the genetic optimization algorithm obtained a better result in less time and iteration. In comparing the two algorithms, the genetic optimization algorithm reduced the standard deviation by 19%, bringing its value to 0.42. On the other hand, the particle swarm optimization algorithm for a longer time, reduced the standard deviation by 14%, bringing its value to 0.44.

The genetic algorithm in optimizing the building roofs obtained excellent results with a total cost of 4678 and a standard deviation of 0.4177. It converged quickly with the 12100 number of objective function evaluations and significantly reduced both the cost function parameters (The genetic algorithm has reached the best possible answer). The particle swarm optimization algorithm also failed to achieve an answer as good as the genetic algorithm with a total cost of 4965, a standard deviation of 0.4430, and a 20100 number of objective function evaluations. About the comparison between these two algorithms, the genetic, with less than 3000 objective function evaluations, was able to experience the most optimal solution that particle swarm algorithm reached with the 20100 number of function evaluations. The use of metaheuristic algorithms in practical problem optimizations, which we frequently encounter in various industries today, can be very efficient. The results of these algorithms are very suitable despite the differences in the outputs, and it will be impossible to reach such answers to different problems without using such algorithms. In future work, based on what we obtained in this research, we suggest using other optimization algorithms or even powerful modeling algorithms such as artificial neural networks. Also, it is possible to study the change in building roof covers and the use of newer coverings in moderating the temperature by adopting new parameters from the cost function in optimization and deep learning algorithms.

The speed of vehicles can affect the performance and damage of the pavement, The two major failures in flexible pavements are fatigue cracks and rutting. The tensile stress under overlay and the compressive stress on the subgrade depend on various factors such as the speed and amount of passing load and contact surface pressure. In this research, machine learning algorithms have been used to predict the critical failure of flexible pavement based on vehicle speed. In the process of using neural networks, the data of neural networks first selects a series of random values as the primary weights and biases of the network which is one of its deficiencies. Therefore, Particle Swarm Optimization (PSO) algorithm is used to optimize the weight of neural networks. In comparison with optimization algorithms, PSO is simpler for implementation and can rapidly find optimal point. According to the modeling results, at low vehicle speeds, the probability of fatigue failures, rutting, and  settling increases, while the lower the speed of the vehicle, the number of repetitions is reduced, and the increase in speed causes that Less damage to the pavement.

Cable-stayed bridges are bridges in which the deck is connected directly to the towers by diagonal cables. These bridges are mainly known by the length of the middle span, which often falls between the span lengths of suspension bridges and bridges without cables. They are also known for their light weight, beauty, and elegance. Cable-stayed bridges are highly indeterminate and are considered significant structures, which makes their optimization a major and costly challenge for designers. The deflection and distribution of the bending moment on the deck and towers depend on the pre-tensioning force applied to the cables, making the calculation of tensile force in cables one of the most critical optimization parameters. Typically, cable pre-tensioning force is calculated through an optimization process that requires extensive analysis, which in turn requires a significant amount of time and computational power. This research aims to reduce the optimization process time by improving the analysis speed. Instead of analyzing a complete three-dimensional model of the structure, this research uses a separate two-dimensional model of the deck and tower without considering the cables. The Lali cable-stayed bridge located in Khuzestan, Iran, serves as the case study for this research. Results show that calculating tensile forces using this method induces an insignificant error compared to optimization using a complete model. The proposed model eliminates the repetitive process of updating the stiffness matrix, significantly reducing the time required to calculate cable pre-tensioning forces during the optimization process. Additionally, the proposed model produces negligible errors both in the optimization process and in the calculation of tensile forces in the cables compared to results obtained through three-dimensional modeling.

Iran is a country that has faced all kinds of natural and unnatural accidents throughout history. Due to its vast geographical area, the need to effectively respond to incidents and accidents in the country is considered one of the basic challenges of crisis managers. Planning to minimize the response time is one of the solutions that can be effective in this field due to the preservation of the golden rescue time. In this research, Red Crescent community rescue and relief bases have been evaluated as one of the elements of responding to incidents and accidents that have inherent and legal duties in crisis management, in terms of how to cover based on the optimal response time. In order to achieve the objectives of the research, which is to identify suitable areas for the development of rescue and rescue centers and to determine the functional radius according to the conditions of the region in the axes and the network of different roads and the time of arrival of the rescue centers to the place of danger, from the method of network analysis. and multi-criteria decision making and particle swarm optimization algorithm (PSO) were used. Also, in order to benefit from it as a practical research, all the desired indicators in the instructions for creating a rescue base have been used to optimize the model. The application of spatial restrictions and advantages has been evaluated using the particle swarm optimization algorithm in the main intercity communication network of the country. The results show that the level of roads covered by rescue services has been increased from 33.61% in the current situation to 55.46%.

There are many factors causing damages to a structure, including earthquakes, winds, environmental effects, etc. In order to repair a damaged structure, first its damage locations should be identified. Therefore, the damage identification of structures is considered as an important issue in civil engineering as well as mechanical engineering. Many methodologies have been devised for damage identification of structures, which are generally categorized to destructive and non-destructive cases. As a non-destructive damage identification approach, solving inverse problems for identifying the properties of a damaged structure is one of the popular methods which utilizes an optimization algorithm to minimize an error function in terms of measured strains or displacements. Since an iterative procedure with significant number of structural analyses should be carried out for the optimization process, an efficient numerical method should be employed to reduce the total computational cost. In this paper, the identification of hole in two-dimensional continuum structures is investigated with finite cell method and particle swarm optimization algorithm. The finite cell method is an efficient numerical method for solving the governing equations of continuum structures having geometrical complexity and/or discontinuities, which uses the concept of virtual domain method. The use of this concept makes the mesh generation easier such that the simple structured meshes can be utilized even for the curved boundaries of a structure, and hence mesh refinement is not necessary for the problems like damage detection. The finite cell method uses adaptive numerical integration for the cells including non-uniform material distribution. Accordingly, quadtree integration is utilized for the structural analysis using the finite cell method. Consequently, the computational time is significantly reduced. On the other hand, particle swarm optimization is a well-known meta-heuristic algorithm, and hence it does not require the gradient information of the problem. This population-based algorithm has been inspired by the social behaviour of animals such as fish schooling and birds flocking. The basis of this algorithm relies on the social influence and learning which enable individuals to preserve cognitive consistency. Thus, the exchange of ideas and interactions between individuals can lead them to solve optimization problems like damage detection. This study proposes the finite cell method and particle swarm optimization algorithm for damage detection of plate structures with single hole or multiple holes. As a non-gradient-based method, particle swarm optimization explores the search space to find the coordinates of the existing damage by minimizing an error function. This error function is evaluated by the strains or displacements calculated by the structural analysis utilizing the finite cell method. In order to evaluate the proposed methodology, numerical examples are provided to demonstrate the capability of finite cell method and particle swarm optimization algorithm in damage detection of two-dimensional structures. The first example considers the damage detection of a plate with a single hole, and it also considers the effects of mesh density. The second example employs a plate structure with three holes. The results demonstrate that the proposed methodology, with suitable computational efforts, can successfully be applied to damage detection of these structures.

One of the most important issues in damage detection is determining the severity of damages. Speed and accuracy in determining the severity of damages depend on the measurement noise. Using the proposed method, damage severity of damaged connections has been calculated in the presence of noise in a faster and more accurate fashion. In the proposed method, instead of using several natural mode shapes, only first proper orthogonal mode shape is used and the damage severity of the rigid connections of moment frames is calculated with acceptable accuracy. Several numerical examples and a laboratory model have been studied to prove the performance of this procedure. The results show that this method has acceptable accuracy and speed to estimate the damage severity in the rigid connections of steel moment frames. The proposed method is based on the finite element model updating method and the objective function is based on the difference between the proper orthogonal decomposition mode shapes of the intact and damaged structures. In the proposed procedure, the wavelet transform is used to reduce the noise in the displacement history of the structure. An optimized image processing method has been used to record structural displacements in the laboratory sample. To optimize the objective function, the particle swarm optimization method has been used. Two numerical scenarios have been studied using the finite element model of a three-story, three-span steel moment frame structure. Also, three laboratory scenarios have been studied on the three-story one-span steel moment frame laboratory model. In this paper, the exact location of the damaged connections is determined using the optimized mode shape curvature difference method. The effects of noise on the accuracy and efficiency of the proposed method are considered. Numerical scenarios have been studied at three noise levels. The proper orthogonal decomposition mode shapes are calculated using the free vibration of the structure under initial displacement.

Considering that the mission of the project organization management is to create a commitment to the timetable, which prevents project delays and increases associated costs, and accept the fact that the delay in the completion time of the project results in the non-economization of the project, Therefore, the necessity of using proper planning and control methods seems necessary. One of the new methods used in project planning and control, which today focuses on many researchers, is the critical chain management method. In this research, multi-project scheduling with resource constraint in the critical-chain model has been investigated. In general, RCMPSP seeks to find the right sequence for project activities, so that the constraints of the priority of the project network and the various types of resource constraints in the project are met simultaneously. To this end, a multi-objective optimization model has been developed to solve this problem, with its stated objectives, the completion time of the entire project, the cost of the overall project implementation, and the quality of the implementation of the project activities. Also, the critical chain approach, as a new method of project management, has been used for project scheduling and particle swarm optimization algorithm for solving proposed model of this research.

Due to structural safety and residential comfort, the vibration control of tall buildings under earthquake and wind excitations has always been one of the important issues in the structural engineering context. One of the well-established approaches for controlling the structural vibration is the use of Tuned Mass Dampers (TMD) employed with different methods in structures. In this paper, a 10-story shear building with linear behavior is studied under twenty eight far-field and near-field earthquakes in MATLAB. Active Tuned Mass Damper (ATMD) is used to control the structural vibration. According to the earthquake excitation random nature, Fuzzy Logic and Mamdani Inference System are applied to determine the control force. In addition, the Particle Swarm Optimization (PSO) algorithm is used to determine the optimum TMD actuator power, and in this optimization, the effect of the actuator saturation is also considered. Furthermore, a method is introduced for robust optimum design of the suggested controller. Using the proposed control system and the optimum actuator power, structural responses decline about 44 pct. Additionally, due to the existence of uncertainty in earthquake records, applying a controller with average actuator power generally results in 33 pct. structural response reduction, and the performance of the active controlled system always outperforms the passive controlled system with 16 pct. structural response reduction.

In the current century, the pattern of competition in production and services has shifted from competition among independent firms to competition among supply chains. Meanwhile, the issue of strategic design of the supply chain is one of the key issues in the field of supply chain management. In these issues, the aim is generally to provide a number of manufacturing facilities and distribution centers at the candidate sites (in different geometrical areas) in order to cover the demand for various products in the customer’s area. In the proposed model, demand for each product in customer areas is achieved through a number of distribution centers, while these distribution centers provide products from manufacturing factories, and these factories must purchase the requirement components to design and produce each product from suppliers. Here, a single-objective mathematical model of mix integer linear programming is proposed that simultaneously considers the facility location problem in addition to the design of the multi-level supply chain. In this model, there are various objectives, such as minimizing the costs of the procurement and production of components, the preparation and production of products, storage and dispatch the products to distribution centers, sending to customer areas and the cost of building factories and distribution centers. To evaluate the proposed model, the mathematical model is first coded with GAMS optimization software, and then solved for large-scale problems with particle swarm optimization algorithm and imperialist competitive algorithm. Statistical analysis of the results indicates the quality of the model and the proposed solving procedure.

One of the issues within the field of transportation is the selection of the transportation investment projects amongst which the selection of the projects dealing with the construction of the new streets or the expansion of the existing streets is of great interest. Although much research has been dedicated to the above-mentioned area, none has solely focused upon the expansion of the streets existing within the urban transportation network. The prime objective of the current study is to select a set of streets within the urban transportation network to enable the expansion purposes (e.g. in terms of budget constraint) as well as to optimize the travel time within the network.To fulfill the aforementioned goal, a mathematical model for the selection of projects within the urban network is proposed for the situations in which merely the expansion of the existing streets is required. Then, an innovative algorithm is introduced to determine the set of candidate streets for expansion purposes. Since the problem of project selection within the urban transportation system is an NP-hard problem from the solution perspective, with one of its most time demanding components being the solution of the traffic assignment sub-problem, and given that the expansion of the existing streets does not change the body and topology of the base network, an innovative algorithm is proposed to solve the traffic assignment problem in this situation, followed by the particle swarm optimization meta heuristic algorithm to solve the project selection model.

Cross-dock is a logistic strategy with the aim of reducing inventory and increasing customer satisfaction.It is described as process of moving goods from suppliers to customers through cross-dock centers without or with little storage, and products must be collected to cross-dock centers before being shipped to customers and after consolidation, freights are delivered to customer with outbound vehicles. This research considers location-routing problem with cross-dock throughout a multi-echelon supply chain by integrating suppliers, cross-dock centers and costumers. In this problem, cross-dock center, vehicle, supplier, costumer, and product are multiple. Also, cross-dock centers and vehicles have limited capacity, split pickup and delivery are allowed, and vehicles are heterogeneous and have constraint for product type shipping. The aim of the model is to minimize the sum of the opening costs of cross-dock centers and the distribution and operational costs of vehicles. This problem arises from the perishable food products distribution industry. For the problem, an integrated mixed integer nonlinear programming model is proposed. Since the problem under study is NP-hard, a particle swarm optimization algorithm is presented to tackle larg-sized problems. Also, a lower bound is developed for the problem. In order to show the performance of the introduced algorithm, different test problems with small, medium and large sizes are provided randomly and solved by the proposed approaches. Computational results show the high performance of the proposed algorithm.

Because of using the different polarization of electromagnetic wave in Polarimetric imagery، it provides a rich source of information from the several aspects of targets. Recently، Polarimetric images as a powerful and efficient tool have been interested to identify the various objects in the complex geographic areas. In order to extracting information، classification of Polarimetric image has an important effect. Support Vector Machines (SVMs) due to their operation based on geometrical characteristics and robustness in high dimensional space، are considered as a suitable case for classification of Polarimetric images. However، the performance of SVMs classifier is strongly influenced by its parameters. Therefore، the optimum values for SVMs parameters should be determined to achieve SVMs classifier with maximum efficiency. Traditional optimization techniques because of computational complexities in the large search space usually trap in local optimum. Thereby، it is inevitable to apply Meta -heuristic Algorithms which performe exploration and exploitation to obtain global optimum. In this paper، the potential of Genetic، Bees and Particle Swarm Optimization (PSO) Algorithms as powerfull techniques in determining the optimum SVMs parameters are evaluated. Comparing the results، demonstrates the superior performance of PSO Algorithm in terms of classification accuracy and speed of convergence.