Earticle

The bat algorithm is a recently proposed meta-heuristic algorithm. Usually in solving the problem of optimization, the position of virtual bats is updated by flying speed, which decreases efficiency of the algorithm and accuracy of the solution. This paper has improved the location update strategy and individual selection strategy of bat algorithm, then puts forward an improved bat algorithm. The algorithm is used to solve three typical reliability-redundancy allocation problems, and the simulation experiment results show that the presented algorithm greatly enhances the computation efficiency, convergence speed and precision of the optimal solution in addressing the problem of reliability redundancy optimization. When compared with the best results reported in the previous literatures, the algorithm achieves a better or equally good optimal solution. It is demonstrated that the proposed algorithm is effective in solving reliability-redundancy allocation problems.

Localization is one of the most important research topics in the wireless sensor network applications. To improve the indoor localization accuracy, the centroid localization algorithm based on Mamdani fuzzy system has been adopted to attain the weight between sensor node and anchor node. This paper proposes a novel optimized input membership function by bat algorithm in fuzzy inference system using the data of received signal strength in real indoor condition. The author has realized the algorithm on Zigbee platform and the experimental comparison on other different centroid localization algorithms indicates that Mamdani fuzzy inference adopting the membership function optimized by bat algorithm renders smaller mean localization errors.

This paper presents a Bat Algorithm (BA) for solving economic emission dispatch (EELD) problem with quadratic fuel function. The BA is a new meta-heuristic algorithm which is a powerful optimal solution search algorithm owing easily selected control parameters and high successful rate as well as high ability for dealing with complex constraints. In addition to minimizing electricity generation fuel cost, emission released into the air from thermal plants is also another main objective needs to be minimized. In order to test the performance of the proposed BA one system with two load cases is employed. The obtained result by the BA compared to that from other methods has revealed that the proposed BA is a very promising meta-heuristic algorithm for solving economic emission load dispatch problem.

Aiming at such shortcomings of traditional Bat algorithm as low precision convergence, premature convergence, slow convergence, an improved BA based on inertia weight factor and Lévy flight (ILBA) has been proposed, which has made two modifications on update equations of bat flying position in BA, using inertia weight factor to keep the flight inertia of bat individual, adaptively adjust the exploitation mechanism of the algorithm in different iteration periods, make the algorithm achieve better convergence precision and altering the strategy about position update of bats from Brownian random walks into Lévy flights strategy to effectively avoid local optimism of the algorithm and guarantee its exploration mechanism while taking advantage of heavy-tailed effect of Lévy flight to speed up the convergence. By means of 4 typical test functions simulation, the results show that ILBA boasts faster convergence and superior optimal performance compared to traditional BA and LBA.

Bat algorithm is an optimization method inspired by the echo-location bats to search in nature, hunt prey behavior, combining multi-agent system and evolution mechanism. To improve the search results of BA algorithm, this paper proposes a gauss perturbation bats optimization algorithm based on simulated annealing (SAGBA). Firstly, the bionic principle, optimization mechanism and characteristics of the bat algorithm are analyzed and the algorithm optimization process are defined; Then the idea of the simulated annealing is put into bat optimization algorithm, and Gaussian disturbance is carried out to some individuals using the bat algorithm and strengthen the ability of the bat algorithm jumping out of local optimal solution. Finally, conduct simulations are respectively compared in 20 typical benchmark test functions among bat optimization algorithm, simulates annealing particle swarm algorithm and SAGBA algorithm. The results show that SAGBA algorithm not only increases the global convergence, but convergence speed and accuracy are better than other two algorithms.

A novel hybrid Bat Algorithm (BA) with the Differential Evolution (DE) strategy using the feasibility-based rules, namely BADE is proposed to deal with the constrained optimization problems. The sound interferences induced by other things are inevitable for the bats which rely on the echolocation to detect and localize the things. Through integration of the DE strategy with BA, the insects’ interferences for the bats can be effectively mimicked by BADE. Moreover, the bats swarm’ mean velocity is simulated as the other bats’ effects on each bat. Having considered the living environments the bats inhabit, the virtual bats can be lifelike. Experiments on some benchmark problems and engineering designs demonstrate that BADE performs more efficient, accurate, and robust than the original BA, DE, and some other optimization methods.

A location fingerprint positioning technology method is proposed based on bat algorithm to improve the defect of traditional indoor location fingerprint positioning technology. These shortcomings include offline heavy workload, limited accuracy and poor robustness. The fingerprint database can generate using the middle point interpolation method and the method that channel attenuation model which instead of offline training stage. This stage realize the function of timing automatic databased updates. After that the database combines K nearest neighbor algorithm with bat algorithm in the stage of matching algorithm to realize the positioning function. Compared with the traditional method, location fingerprint positioning technology using bat algorithm reduces the overall positioning of workload and rapidly respond to the effect of the changing environment. Finally, localization performance test is carried out under the given simulation environment. The result indicates that this method improves the average precision than other algorithms about 23.14%, the vast most of the blind node position error range within 1.5 meters, which shows the advantages of positioning accuracy, robustness and adaptation to the changing environment.