An improved Mamdani Fuzzy Neural Networks Based on PSO Algorithm and New Parameter Optimization

Meng, Liang; Yin, Shoulin; Hu, Xinyuan

doi:10.11591/ijeecs.v1.i1.pp201-206

Cited by 6 publications

(3 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because the existing ABC algorithms [11][12][13] have the defect of restricting the escape of precocious individual [13]. We deign a new escape scouter strategy.…”

Section: Escape Scouter Strategymentioning

confidence: 99%

An Improved Artificial Bee Colony Algorithm for Staged Search

2016

Self Cite

View full text Add to dashboard Cite

Artificial Bee Colony (ABC) Keywords: artificial bee colony, staged search, function optimization, escape radius, uniform distribution, backward learningCopyright © 2016 Universitas Ahmad Dahlan. All rights reserved. IntroductionAs we all know, Artificial Bee Colony (ABC) [1,2] is an intelligent optimization algorithms imitating bees foraging in the nature, which has the character of easily implement and setting parameters. State-of-the art of the field the report is that artificial bee colony can be used for anomaly-based intrusion detection systems. Also an artificial bee colony algorithm is presented for data collection path planning in sparse wireless sensor networks. ABC algorithm is used for function optimization and combines genetic algorithm, particle swarm algorithm and finite difference algorithm to solve some complex problems especially TSP problems [3,4]. ABC algorithm also is used in neural network training and digital IIR filter designing.However, in the practical engineering application, many production practice problems are transformed into high dimensional complex function optimization problems. But it has characters of function complex, great scale, high dimensions and nonlinear. When we use classical optimization method to solve this question, it is easy to fall into local extremum with the increase of dimension. ABC exists some disadvantages, such as premature convergence, easy to fall into local extremum and low solution. So Mansouri [5] proposed a novel iterative method combining ABC and Bisection method to find the fixed point of a nonlinear function effectively. Imanian [6] proposed a modified ABC algorithm called VABC to overcome this insufficiency by applying a new search equation in the onlooker phase. It used the PSO search strategy to guide the search for candidate solutions. Wang [7] proposed a novel multi-strategy ensemble ABC algorithm. A pool of distinct solution search strategies coexisted throughout the search process and competed to produce offspring. Mustafa [8] proposed integration of multiple solution update rules with ABC which used five search strategies and counters to update the solutions. P. Mustafa [9] added directional information to ABC algorithms. The new scheme was compared with basic ABC and ABCs with MR. It examined the performance of this method on well-known nine numerical benchmark functions to show it effectively.The above improved ABC algorithms achieve good effect, But they also has some disadvantages with more control parameters, premature convergence et al. To solve those

show abstract

“…Because the existing ABC algorithms [11][12][13] have the defect of restricting the escape of precocious individual [13]. We deign a new escape scouter strategy.…”

Section: Escape Scouter Strategymentioning

confidence: 99%

An Improved Artificial Bee Colony Algorithm for Staged Search

2016

Self Cite

View full text Add to dashboard Cite

show abstract

“…Therefore, we compare TVAC with our new method (EM-CMLSQN), and the simulation results show that EM-CMLSQN has a better convergence rate and performance of jumping out of local solution than PSO and TVAC method. We also apply EMCMLSQN algorithm into path planning problem, and the results represent that EM-CMLSQN algorithm can search the optimal path more precisely and can be better applied into solving discrete domain problems than genetic algorithm and PSO algorithm [10,11]. The basic EM algorithm is composed of initialization, local search, resultant force calculation, particle displacement and judgment terminated.…”

Section: Introductionmentioning

confidence: 99%

Motorcycle-Security using Position Searching Algorithm Based on Hybrid Fuzzy-Dijkstra

Mubarok

Firmansyah

Haryanto

et al. 2016

IJEECS

View full text Add to dashboard Cite

As we all know, traditional electromagnetism mechanism (EM) algorithm has the disadvantage with low solution precision, lack of mining ability and easily falling into precocity. This paper proposes a new chaos electromagnetism mechanism algorithm combining chaotic mapping with limited storage QuasiNewton Method (EM-CMLSQN). Its main idea is that it adopts limit quasi-Newton Key words: electromagnetism mechanism, limited storage quasi-newton method, limit quasi-newton operator, chaos mapping, particle swarm optimization, discrete artificial bee colony algorithmCopyright © 2016 Institute of Advanced Engineering and Science. All rights reserved. Introduction.Electromagnetism Mechanism (EM) algorithm [1, 2] is a kind of random search methods based on principle of attraction and repulsion between different charged particles in Coulomb law and electromagnetic field. This algorithm firstly establishes the relationship of fitness function value and individual's value affected by electric field, then it is as the population movement trend according to the principle of "excellent solutions attract poor, poor solutions reject the excellent" and puts forward a global stochastic optimization heuristic algorithm [3]. The global stochastic optimization heuristic algorithm has been used in many aspects, such as fault location in distribution networks and pipeline assemble. Nevertheless, the EM algorithm has the shortcoming of premature convergence, low local search accuracy in late and slow convergence rate, which resembles other global intelligent algorithms. To solve these problems, this paper proposes an improved local optimization strategy using high precision local optimization operator and limited storage Quasi-Newton operator [4]. It seeks optimization value for solution domain near optimal individual and adopts chaos mapping [5] to increase the diversity of population. As a mature intelligent algorithm, Particle Swarm Optimization (PSO) algorithm [6,7] has a very good effect on searching the optimization value in continuous domain. There is an improved PSO algorithm named particle swarm optimization with TimeVarying Accelerator Coefficients (TVAC) [8,9], which has a better ability of searching optimization. Therefore, we compare TVAC with our new method (EM-CMLSQN), and the simulation results show that EM-CMLSQN has a better convergence rate and performance of jumping out of local solution than PSO and TVAC method. We also apply EMCMLSQN algorithm into path planning problem, and the results represent that EM-CMLSQN algorithm can search the optimal path more precisely and can be better applied into solving discrete domain problems than genetic algorithm and PSO algorithm [10,11].

show abstract

“…The accuracy of speaker recognition is needed by system recognition, where the recognition system must be able to perform an accurate response in accordance with data from the human speech [2]. Artificial intelligent approaches have been used to increase accuracy in speaker recognition; one of the methods in the artificial intelligent approach is the fuzzy Mamdani method [3]. The Mamdani method can be used to identify a non-linear system such as speaker recognition [4].…”

Section: Introductionmentioning

confidence: 99%

Speaker Recognition in Content-based Image Retrieval for a High Degree of Accuracy

et al. 2018

View full text Add to dashboard Cite

The purpose of this research is to measure the speaker recognition accuracy in Content-Based Image Retrieval. To support research in speaker recognition accuracy, we use two approaches for recognition system: identification and verification, an identification using fuzzy Mamdani, a verification using Manhattan distance. The test results in this research. The best of distance mean is size 32x32. The best of the verification for distance rate is 965, and the speaker recognition system has a standard error of 5% and the system accuracy is 95%. From these results, we find that there is an increase in accuracy of almost 2.5%. This is due to a combination of two approaches so the system can add to the accuracy of speaker recognition.

show abstract

An improved Mamdani Fuzzy Neural Networks Based on PSO Algorithm and New Parameter Optimization

Cited by 6 publications

References 7 publications

An Improved Artificial Bee Colony Algorithm for Staged Search

An Improved Artificial Bee Colony Algorithm for Staged Search

Motorcycle-Security using Position Searching Algorithm Based on Hybrid Fuzzy-Dijkstra

Speaker Recognition in Content-based Image Retrieval for a High Degree of Accuracy

Contact Info

Product

Resources

About