artificial bee colony algorithm

Increasing the depth of microchannels on metallic bipolar plates (BPPs) in PEM fuel cells leads to an increase in the efficiency. In this research, the stamping process has been applied for manufacturing of the BPPs made of commercially pure titanium with a direct parallel flow field. The effect of process parameters including die clearance, forming speed, and sheet/die friction coefficient on the filling rate and thinning of the BPPs was investigated. The required tests were designed via the response surface method (RSM), implemented by a validated finite elements (FE) model, and the desired outputs were extracted. Then, a quadratic equation was presented for predicting the filling rate based on the input parameters using the regression method. In the following, using the artificial bee colony algorithm, the coefficients of the mentioned equation were enhanced and its error was decreased almost by 53%. Finally, an artificial neural network (ANN) was used to predict the filling rate. The results demonstrated that the proposed ANN model is very effective and approximates the filling rate of the microchannel with high accuracy.

Estimation  of  the  load  carrying  capacity  of  pile  foundation  is  one  of  the  most  sought  after  research  areas  in geotechnical   engineering.   Static   equilibrium   and   other dynamic equations are used to predict  the  axial  load  capacity  of  pile.  The  prediction  of lateral  load  capacity  of  piles,  used  in  tall  and  offshore structures is more complex and requires solution of non-linear differential equations. The elastic analysis adopting Winkler   soil   model is   not suitable for the non-linear soil behavior.  Estimating the load capacity of such piles using experimental methods is always associated with error and makes the modeling result far from reality. Today, intelligent methods have shown a high capability in predicting and estimating unknown variables and can replace experimental and analytical methods. In this research, we tried to accurately predict the lateral load capacity of piles in clay soils by creating an intelligent hybrid model called optimized relevant vector regression with the artificial bee colony algorithm. The relevant vector regression is a probabilistic method based on Bayesian approach. The relevant vector regression does not need to predict the error/margin tradeoff parameter C, which can decrease the time and the kernel function, does not need to satisfy the Mercer condition. For those relevant vector regression advantages compared with the support vector regression approach, relevant vector regression model is successfully applied in regression prediction problems. In this method, relevant vector regression is used as a predictive model and artificial bee colony algorithm is used to optimize the parameters of relevant vector regression method. The artificial bee colony algorithm is a swarm based meta-heuristic algorithm for optimizing numerical problems. It was inspired by the intelligent foraging behavior of honey bees. The algorithm is specifically based on the model for the foraging behavior of honey bee colonies. The model consists of three essential components: employed and unemployed foraging bees, and food sources. The first two components, employed and unemployed foraging bees, search for rich food sources, which is the third component, close to their hive. The model also defines two leading modes of behavior which are necessary for self-organizing and collective intelligence: recruitment of foragers to rich food sources resulting in positive feedback and abandonment of poor sources by foragers causing negative feedback.  In artificial bee colony, a colony of artificial forager bees (agents) search for rich artificial food sources (good solutions for a given problem). To apply artificial bee colony, the considered optimization problem is first converted to the problem of finding the best parameter vector which minimizes an objective function. Then, the artificial bees randomly discover a population of initial solution vectors and then iteratively improve them by employing the strategies: moving towards better solutions by means of a neighbor search mechanism while abandoning poor solutions.In this modeling, the data used are related to a laboratory data set of small-scale pile load capacity. Various statistical indicators were used to evaluate the modeling accuracy. Finally, the results showed that the combined relevant vector regression with the artificial bee colony algorithm for test data with R2 = 0.975 and RMSE = 0.001, has a high ability to predict the lateral load capacity of spark plugs. In addition, the sensitivity analysis performed in this study showed that the variables of eccentricity of load and the length of pile are more important and effective compared to other parameters.

In this research, the dual resource constrained flexible job-shop scheduling problem (DRCFJSP) is considered. Compared to the flexible job-shop scheduling, there is a limited research on DRCFJSP. The flexible job-shop scheduling problem is an extension of the classical job-shop scheduling problem by allowing an operation to be assigned to one of a set of eligible machines during scheduling. Hence, solving DRCFJSP not only needs to determine the processing sequences on machines and assign each operation to a machine, but also needs to determine a worker among a set of skilled workers for processing operation on the selected machine. The problem in this study was investigated to minimize two objectives consisting of total weighted tardiness and maximum completion time. The lexicographic approach is applied to compare the solutions and select the optimum solution. The first objective function is total weighted tardiness. DRCFJSP is strongly NP-hard, so a hybrid artificial bee colony algorithm is proposed to solve medium and large instances. In order to evaluate the performance of the proposed algorithm, computational studies have been conducted and compared with the results of the GAMS software. The results show that proposed hybrid algorithm has an appropriate performance for solving the DRCFJSP.

In recent years, the damage identification of structures becomes more attractive for researchers in order to assess ‎the quantify condition of structural system during service life. Moreover, identifying the damage location and ‎severity is very important after disaster such as earthquake and terrorist attak. Structures can be also damaged by ‎normal activity such as corrosion, aging, fatique, wind, waveload etc. Therefore the structural health monitoring is ‎an emerging field to ensure the continues and periodic performance of structures. In this paper, identification of the ‎extent and location of damages in structures are studied by analytical method using artificial bee colony ‎optimization (ABC). In the analytical method, the mass and stiffness matrices of structure could be determine by ‎the finite element procedure. Considering the stiffness matrix of healthy structure and that of the damage structure, ‎the location and severity of the damage could be determined. It is assumed that the global mass matrix remains ‎unchanged after the damage occures in the structure. The natural frequencies and mode shapes of damaged ‎structure can be obtained by measurement. In the study, the damage characteristics are known. Then by applying ‎the eigenvalue equation, the stiffness matrix is determined for damaged structure. Finding the extent and location ‎of damage is introduced as an inverse problem. Using the conventional methods are very expensive and time ‎consuming, while metaheuristic evolutionary computing method is capable to solve complex combinational ‎optimization problems. Swarm intelligence algorithm introduces the collactive behavior of social insects colonies ‎to solve optimization problems. Artificial bee colony algoritm is an evolutionary computing method, which is ‎developed, based on the intelligent foraging behavior of honeybee swarm. Each food source is considered as a ‎possible solution. The location and quality of the nectar from the flower is related to the damage properties and ‎fitness function, respectively. The dimension of every artificial employed bee is equal to the number of member of ‎the structure. Then quality value of the food source is evaluated by the fitness function. The best fitness value is ‎memorized in each search. When the fitness value denote improved after a predefined iterative, the new possible ‎solution will be considered. In the ABC process, the number of food source, the limit and the maximum cycle ‎number are three control parameters. In the optimization problem, applying a proper objective function is one of ‎the indispesable part of the process. Since the structural damage detection is a highly nonlinear problem, a proper ‎objective function can detect the damage accurately and quackly. There are various methods for damage detection, ‎which generaly can be classified into two categories, static and dynamic method. Because of the efficiency of the ‎dynamic method, the objective function is selected based on the dynamic technique, which utilize the eigenvalue ‎problem. In the mathematical equation of the objective function, the mass and stiffness matrix of healthy structure ‎is defined by finite element method. The natural frequencies and mode shapes obtained by the measurement. The ‎stiffness matrix of damaged structure is determined with the optimization algorithm to minimize the objective ‎function. In a measurement test, the used sensors cannot detect all of the degrees freedom of a structure, therefore ‎the obtained information in measurement include a limited number of frequencies or mode shapes. In addition, to ‎avoid a time consuming process, it may be decided to utilize only a limit number of frequencies obtained by the ‎measurement. The system equivalent reduction expansion process (SEREP), which is an accurate and efficient ‎technique of model reduction, is utilized in the paper. Moreover, the damage detection is examined through three ‎numerical examples, plane and space truss and palne frame, each one has two damage scenarios, which include ‎noisy measurement data. The results indicate that the proposed method is a powerfull procedure to detect damages ‎in structures.‎

Intrusion in the network is increasing. Intrusion detection system can greatly prevent network attacks. Feature selection is a critical issue in intrusion detection systems which have a considerable impact on the accuracy and effectiveness of the system. In this study, a new hybrid network intrusion detection system with improved artificial bee colony algorithm using support vector machine classifier is proposed for feature selection. The main idea is utilizing a combination of search equations of particle swarm optimization and Differential Evolution for updating bee’s position of employed and onlooker bees and utilizing levy flight on scout bees phase, to improve exploitation and increase the convergence rate of the standard artificial bee colony algorithm. The robustness and stability of the proposed approach is evaluated on NSL-KDD dataset and showed significant improvement on the overall performance of intrusion detection system with an accuracy of 98.97 percent.

In the design of seismically base-isolated structures, it is expected that the isolator will experience nonlinear behavior while the superstructure still behaves linearly. Therefore for modeling these systems, a linear behavior is assumed for superstructure and different nonlinear models are used for isolator. But there are special conditions such as strong ground motions in which superstructure behave nonlinearly. In this study, the nonlinear behavior of seismically base-isolated structures is more accurately investigated. This is done using nonlinear time history analysis of structures using ground motions. Two sets of ground motions are selected which represent earthquakes with 475 and 2475 years return period. OpenSees software is used for modeling these structures. The effective parameters on the response of seismically base isolated structure which are investigated are: response modification factor of the superstructure, stiffness of the isolator, damping ratio of the isolator, stiffness of the superstructure and damping ratio of the superstructure. Studies of this paper are divided into two parts. In the first part, two-degree freedom model with viscoelastic isolator has been used to investigate the effect of superstructure nonlinearity. Also a sensitivity analysis is done to find important parameters which have more effects on the systems response. Results of this part show that, nonlinear behavior of superstructure increases system ductility demand drastically. It is concluded that the period of isolator and superstructure have the most effect on the ductility demand. In the second part, the effect of different parameters and higher mode effects on the response of seismically base-isolated structures is investigated using muti-degree of freedom models sited on isolator with bilinear behavior. Results obtained in this part also confirm the increase in the response of the system when the superstructure has low strength. Likewise in this condition, the isolator deformation decreases. Distribution of ductility demand in the height of structure is also non-uniform in this condition and lower stories are more vulnerable. Isolators with a lower fundamental period and also isolators with a lower yield force lead to the least amount of isolator deformation and ductility demand of superstructure. By increasing damping in the isolator, ductility demand of superstructure will increase. A stiffer superstructure with nonlinear behavior has a much more ductility demand rather than similar structure which is more flexible. But when the superstructure behaves linearly, the fundamental period of superstructure and isolator deformation increase or decrease together. Finally, due to the intensive increase in the ductility demand of the superstructure when it behaves nonlinearly, a solution is proposed with the aim of simultaneous optimization of the superstructure and isolator systems. This strategy uses a combination of uniform ductility theory and a modified artificial bee colony algorithm. By applying this method and using ductilities of superstructure and isolator simultaneously, it is possible to obtain the desirable ductility in the isolator, in addition to achieve a more economical choice for the superstructure. If there is no constraint around the structure for movement, by increasing the target ductility of the superstructure, the weight of the superstructure can be reduced by reducing its yield force.

Artificial Bee Colony Algorithm(ABC) is one of the swarm intelligence optimization algorithms that is extensively used for the goals and applications static. Many practical, real-world applications, nevertheless, are dynamic. Thus we need to get used optimization algorithms that could be solved problems in dynamic environments as well. Dynamic optimization problems where change(s) may occur through the time. In this paper we proposed one approach based on chaotic ABC combined with explicit memory method, for solving dynamic optimization problems. In this proposed algorithm, we used the explicit memory for store the aging best solution for the maintaining diversity in the population. Use the aging best solution and diversity in environments helps the speed convergence in algorithm. The proposed approaches have been tested on Moving Peaks Benchmark. The Moving Peaks Benchmark is the suitable function for testing optimization algorithms in dynamic environments. The experimental study on a Moving Peaks Benchmark show that proposed approach has a superior performance in comparison with several other algorithms in dynamic environments.

The development of effective decision support tools that can be adopted in the transportation industry is vital since it can lead to substantial cost reduction and efficient resource consumption. However, vehicles moving on our roads contribute to congestion, noise¡ pollution, and accidents. So route planning and transport management, using optimization tools, can help reduce transport costs by cutting mileage and improving driver and vehicle usage. In addition, it can improve customer service, cut carbon emissions, improve strategic decision making and reduce administration costs.
Due to the simultaneous pick-up and delivery postal service and delivery time importance of those parcels, this study focuses on the pick-up and delivery problems. The pick-up and delivery problems are important types of vehicle routing problem (VRP). VRP is the core of scientific research on the distribution and transport of people and goods. Unlike the classical VRP, in which all customers require the same services, in the pick-up and delivery problem basic it is considered that two different types of services can be found in one place, in fact there's a pick up or delivery. PDP has several applications in the transportation of pick-up and delivery parcel post. The purpose of this research is to find the most optimal route to transport postal service. It is performed by imposing a series of conditions to the pick-up and delivery problems using meta-heuristic algorithms for the simulation data. It is followed by brief explanation of present metaheuristic algorithms including bee colony algorithm and genetic algorithms and their features. Finally the results of the algorithms are compared on the basis of the accuracy, repeatability, speed of convergence. It is necessary to note that the results are not ideal, but the best case is considered. The results showed the performance of the bee algorithm are better than genetic. Based on the results obtained in each run, genetic algorithms and Bee were 84% and 93% are possible to achieve the best solution.

In mathematics and computer science an optimization problem, the problem is finding the best solution among all possible solutions. Given the importance of the knapsack in computer sciences, different algorithms are used to solve it. Knapsack problem is a combinational problem of selectivity and the purpose of solving the most benefit by taking the capacity is the tolerable knapsack. Since the knapsack is a problem of constrained maximization. In this study, a mathematical model in the form of a function unlimited minimization and designed for it, hen this model on Particle Swarm Optimization , Firefly Algorithm and Artificial Bee Colony has been implemented in MATLAB software environment, The results show that the artificial bee colony algorithm, the model is better than the other two algorithms .The advantage of this model is the objective function , because minimization and unlimited models , to implement with many Bio-Inspired algorithms.

To calculate shielding failure of a lightning protection system, several analytical methods such as Electro-Geometric Model and olling Sphere Method have been proposed. These methods are used to determine protected and unprotected areas, but they cant show the frequency of shielding failure to determine more risky areas. In this paper, a new numerical model is proposed by using eader Progression Model and Artificial ee olony Algorithm to identify the most vulnerable area on complex structures. The progression of lightning downward leader, the inception and propagation of upward leaders are considered step by step until final jump. Structures, air terminals, downward and upward leaders are modeled by different shapes of charges, and the electric field in an environment is computed by charge simulation method. ightning is assumed as artificial bee, and areas with high shielding failure are the best food sources with high nectar amount. So the result produced by A algorithm is the most lightning risky area. Finally, lightning performance of an asymmetric structure protection system is evaluated, and based on simulation results the best position of air terminals is determined. The proposed method can be used for protection system design of any complex structure.