Genetically Optimized Multi-Layer Fuzzy Polynomial Neural Networks: Analysis and Design

2006 ◽  
Vol 10 (1) ◽  
pp. 35-49 ◽  
Author(s):  
Sung-Kwun Oh ◽  
◽  
Witold Pedrycz ◽  
Ho-Sung Park ◽  
◽  
...  
Keyword(s):  
Neural Networks ◽  
Design Procedure ◽  
Least Square Method ◽  
Least Square ◽  
Genetic Optimization ◽  
Analysis And Design ◽  
Networks Analysis ◽  
The One ◽  
Neurofuzzy Networks ◽  
Selection Of

In this study, we introduce a new category of neurofuzzy networks – Fuzzy Polynomial Neural Networks and develop a comprehensive design methodology involving mechanisms of genetic optimization, and genetic algorithms, in particular. The augmented genetically optimized FPNN (referred to as gFPNN) is a structurally optimized architecture which comes with a higher level of flexibility in comparison to the one we have encountered in the conventional FPNN. The GA-based design procedure being applied to each layer of FPNN leads to the selection of preferred nodes (or FPNs) available within the FPNN. In the sequel, two general optimization mechanisms are explored. First, the structural optimization is realized via GAs whereas for the ensuing detailed parametric optimization is carried out in the setting of a standard least square method-based learning. The performance of the gFPNN is quantified through experimentation where we use a number of modeling benchmarks – synthetic and experimental data already experimented with in fuzzy or neurofuzzy modeling.

Parameter sensitivity analysis and optimum model of the magnetorheological damper’s Bouc–Wen model

2020 ◽  
pp. 107754632095929
Author(s):  
Min Jiang ◽  
Xiaoting Rui ◽  
Wei Zhu ◽  
Fufeng Yang ◽  
Hongtao Zhu ◽  
...  
Keyword(s):  
Parameter Identification ◽  
Fitness Function ◽  
Least Square Method ◽  
Least Square ◽  
Identification Process ◽  
Long Time ◽  
Method Optimization ◽  
Optimum Model ◽  
The One ◽  
Sensitive Parameters

To overcome the shortcomings of the Bouc–Wen model, such as too many parameters, complex identification process, and long time consuming, the sensitivity of parameters was analyzed. A Bouc–Wen optimum model with sensitive parameters to guarantee calculating accuracy was established. First, according to the results of the magnetorheological damper’s mechanical property test, the sensitivity of Bouc–Wen model’s parameters was analyzed by the one-at-a-time method. Optimization of the Bouc–Wen model was completed. Second, the parameters of the Bouc–Wen optimum model were identified under three harmonic excitations. Compared with the original Bouc–Wen model, the differences of calculation accuracy were 0.0055, 0.0007, and 0.0070 respectively. And the convergence rate of the fitness function for parameter identification increased by 67.89%, 49.94%, and 67.24%, respectively. And the iteration time of 1000 iterations was shortened by 36.52%, 25.95%, and 64.11%, respectively. It indicates that the Bouc–Wen optimum model had higher efficiency and certain accuracy in parameter identification process. Then, the calculation accuracy of Bouc–Wen optimum model with independent and coupled mean parameters were analyzed respectively. Finally, the parameters of the Bouc–Wen optimum model and current were fitted by the least square method. The results showed that the Bouc–Wen optimum model can accurately and efficiently simulate the dynamic characteristics of magnetorheological dampers.

A New Type of Hybrid Learning Algorithm for Three-Layered Feed-Forward Neural Networks

2014 ◽  
Vol 1030-1032 ◽  
pp. 1627-1632
Author(s):  
Yun Jun Yu ◽  
Sui Peng ◽  
Zhi Chuan Wu ◽  
Peng Liang He
Keyword(s):  
Neural Networks ◽  
Learning Algorithm ◽  
Least Square Method ◽  
Hybrid Learning ◽  
Least Square ◽  
Feed Forward ◽  
Hybrid Learning Algorithm ◽  
Layer Function ◽  
Feed Forward Neural Networks ◽  
New Type

The problem of local minimum cannot be avoided when it comes to nonlinear optimization in the learning algorithm of neural network parameters, and the larger the optimization space is, the more obvious the problem becomes. This paper proposes a new type of hybrid learning algorithm for three-layered feed-forward neural networks. This algorithm is based on three-layered feed-forward neural networks with output layer function, namely linear function, combining a quasi Newton algorithm with adaptive decoupled step and momentum (QNADSM) and iterative least square method to export. Simulation proves that this hybrid algorithm has strong self-adaptive capability, small calculation amount and fast convergence speed. It is an effective engineer practical algorithm.

Comparison between CNN and Haar classifiers for surgical instrumentation classification

2017 ◽  
Vol 10 (28) ◽  
pp. 1351-1363 ◽  
Author(s):  
Paula C. Useche Murillo ◽  
Robinson Jimenez Moreno ◽  
Javier O. Pinzon Arenas
Keyword(s):  
Neural Networks ◽  
Object Recognition ◽  
Convolutional Neural Networks ◽  
Video Sequence ◽  
Surgical Instrumentation ◽  
Detection Stage ◽  
Scene Information ◽  
The One ◽  
Selection Of

The following paper presents the development, operation and comparison of two methods of object recognition trained for the classification of surgical instrumentation, where a video sequence is used to capture scene information constantly, in order to allow the selection of some of the instruments according to the needs of the doctor. The methods used were Convolutional Neural Networks (CNN) and Haar classifiers, where the first was added a previous element detection stage, and the second one was conditioned to allow it not only to detect elements, but also to classify them. With the CNN an accuracy of 96.4% in the classification of the two categories of the first branch of the tree was reached, while for Haar classifiers 90% accuracy was achieved in the detection of one of the five instruments, whose classifier was the one that presented the best results.

SIMPLIFIED FUZZY INFERENCE RULE-BASED GENETICALLY OPTIMIZED HYBRID FUZZY NEURAL NETWORKS

2008 ◽  
Vol 16 (02) ◽  
pp. 245-274 ◽  
Author(s):  
BYOUNG-JUN PARK ◽  
WITOLD PEDRYCZ ◽  
SUNG-KWUN OH
Keyword(s):  
Neural Networks ◽  
Fuzzy Inference ◽  
Inference Rule ◽  
Least Square Method ◽  
Fuzzy Neural Networks ◽  
Least Square ◽  
Rule Based ◽  
Fuzzy Partition ◽  
Fuzzy Neural ◽  
Input Variables

In this study, we introduce an advanced architecture of genetically optimized Hybrid Fuzzy Neural Networks (gHFNN) and develop a comprehensive design methodology supporting their construction. A series of numeric experiments is included to illustrate the performance of the networks. The construction of gHFNN exploits fundamental technologies of Computational Intelligence (CI), namely fuzzy sets, neural networks, and genetic algorithms (GAs). The architecture of the gHFNNs results from a synergistic usage of the genetic optimization-driven hybrid system generated by combining Fuzzy Neural Networks (FNN) with Polynomial Neural Networks (PNN). In this tandem, a FNN supports the formation of the condition part of the rule-based structure of the gHFNN. The conclusion part of the gHFNN is designed using PNNs. We distinguish between two types of the simplified fuzzy inference rule-based FNN structures showing how this taxonomy depends upon the type of a fuzzy partition of input variables. As to the conclusion part of the gHFNN, the development of the PNN dwells on two general optimization mechanisms: the structural optimization is realized via GAs whereas in case of the parametric optimization we proceed with a standard least square method-based learning. To evaluate the performance of the gHFNN, we experimented with three representative numerical examples. A comparative analysis demonstrates that the proposed gHFNN come with higher accuracy as well as superb predictive capabilities when compared with other neurofuzzy models.

scholarly journals EVALUATION OF THE RATIONALITY OF ENTERPRISE POTENTIAL MANAGEMENT

2020 ◽  
Author(s):  
A. Shchebel
Keyword(s):  
Neural Networks ◽  
Regression Model ◽  
Capacity Management ◽  
The Other ◽  
Temporal Dimension ◽  
Cost Measurement ◽  
The Difference ◽  
The One ◽  
The Cost ◽  
Selection Of

The potential of the enterprise may have a number of components that are heterogeneous in their economic and managerial nature. This requires the selection of criteria that would be common to assess all components of capacity. From the standpoint of the resource approach, such criteria could be the cost of resources to build the potential of the enterprise, as well as the value created by using existing capacity. This criterion is easily consistent with the goals of formation and implementation of the potential of the enterprise, and therefore can have a quantitative and temporal dimension of achievement. The cost of resources is a cost measurement of the criterion. In turn, time, on the one hand, is one of the dimensions of the selected criterion, and, on the other hand, a separate criterion. After all, the same result, which is obtained for different periods of time, usually has a different assessment. It is substantiated that the assessment of the rationality of enterprise capacity management should be carried out on the basis of comparing the cost of resources that were involved in the formation of potential with the value created as a result of its use. It is proved that the significance of the difference between these values depends on the time factor. Reducing the analyzed period and increasing the difference between the studied values increases the rationality of management. Applying the provisions of the theory of neural networks, a regression model is constructed, which assumes the use of a recurrent function. This ensured the accuracy of forecasting the resulting parameters and increased the informativeness and objectivity of the proposed method.

scholarly journals The simple method of dynamic visco-elastic analysis of road structure. on rheological foundation

2009 ◽  
Vol 4 (1) ◽  
pp. 073-083
Author(s):  
Sławomir Karaś ◽  
Magdalena Sawecka
Keyword(s):  
Dynamic Analysis ◽  
Elastic Constants ◽  
Deformation Process ◽  
Least Square Method ◽  
Least Square ◽  
Elastic Analysis ◽  
Simple Method ◽  
One Dimensional ◽  
Road Pavement ◽  
The One

In contrast to computationally advanced methods of road pavement dynamic analysis, the one-dimensional, simple method is derived on the basis of visco-elastic simple beam lying on generalized Winkler visco-elastic foundation. By virtue of least square method the visco-elastic constants could be estimated with technically admissible accuracy. The introduced method is useful enough to predict any pavement deformation process in the range of linear visco-elasticity.

Sign in / Sign up

Export Citation Format

Share Document