High order -planes integration: A new approach to computational cost reduction of General Type-2 Fuzzy Systems
Introduction
Type-2 Fuzzy Sets were originally proposed by Zadeh and later studied in more detail by Mendel and John (2002) and the conceptual difference of Type-2 Fuzzy Sets with respect to Type-1 Fuzzy Sets is the inclusion of a new secondary domain related with handling uncertainty. The emergence of Generalized Type-2 FS provides a new ability for handling uncertainty in modeling additionally to the vagueness modeling provided by the Type-1 FS. These properties of the FSs allow their application in many kinds of problems, for example, control systems Schouten et al. (2002), Castillo and Melin (2014), Caraveo et al. (2016), Cervantes and Castillo (2015), image processing Melin et al. (2010), Gonzalez et al. (2015), diagnosis systems Luo et al. (2010), Nilashi et al. (2017), classification, industrial problems Hannan et al. (2015), Wati (2016), etc.
It has been shown in previous works that the performance of an IT2 FS improves on the results obtained with T1 FS, and some researchers that have reported this are in Castillo et al. (2016), however, the uncertainty modeling provided by an IT2 FS is incomplete, because the uncertainty is not uniform as assumed in this model. On the other hand, there exist representations that allow the modeling of non-uniform uncertainty, that correspond to approximations of GT2 FS, and these approaches have the potential to overcome the performance obtained by IT2 FS and they are, for example, the Geometric approach (Coupland and John, 2008), the Z-Slices method Wagner and Hagras (2011), Wagner and Hagras (2010), and the -planes approximation (Mendel et al., 2009b).
The different representations have advantages and disadvantages and have been implemented successfully in many applications, however, the main problem of the implementation of the mentioned representations of GT2 FS is the computational cost, so, the aim of the present work has been focused on reducing the computational cost of the -planes representation of GT2 FS.
Then, considering that the -planes representation is based on a discretization of the GT2 FS, it is expected that the approximation accuracy would be proportional to the discretization level. However, the computational cost is also proportional to the discretization, considering that each slice of GT2 FS for the discretization represents an IT2 FS (also called -plane), and this represents a problem for applications that require a good approximation of the theoretical model of GT2 FS.
The proposed approach, then, aims at reducing the number of -planes that approximate the GT2 FS, and this reduction can be possible with the implementation of High-order -planes integration. This is because the conventional method for -planes integration can be expressed as a numerical integrator of order one, then, the proposal is that with the implementation of High-order integration, the number of planes required for a good approximation can be significantly reduced.
To test the improvement with the proposed approach a set of experiments with random GT2 FS composed of different kinds of GT2 MFs were performed. In this case, 30 experiments and a statistical test were performed to demonstrate that the proposed approach exhibits an improvement with respect to the conventional method.
The structure of the present paper is as follows: in Section 2 the basics of Type 2 FS are presented, Section 3 outlines the GT2 MFs used in the present work, Section 4 explains the Closed Newton–Cotes rules, in Section 5 the proposed approach of -planes integrations is presented, Section 6 contains the experimentation and the obtained results, and finally Section 7 offers the conclusion and the future work.
Section snippets
Type-2 Fuzzy systems basics
In this section, the basic concepts of Type-2 Fuzzy Logic, the uncertainty model considered in Type-2 Fuzzy logic and its implications in the computational cost, and in addition the -planes representation of GT2 FIS are presented.
Generalized Type 2 MF
In this section, the GT2 MFs used for the proposed method are introduced. In the present work, we propose the use of symmetrical GT2 MFs, the mathematical definition of a GT2 MF depends on its FOU, and is defined based on an IT2 MF.
Closed Newton–Cotes quadrature rules
The Newton–Cotesrules are high-order numerical integration methods Simos (2008), Venkateshan and Swaminathan (2014), and a brief introduction to these methods is presented as follows. Let us start by considering a one-dimensional integral (15).
To obtain the Newton–Cotes rules, the function range is divided into equal intervals, where (16) is the space between each sample with respect to the next (17).
Based on this discretization, the
High order -plane integration
The proposal consists on realizing the -planes integration, which is a sub process of Type-Reduction of the GT2 FIS, with the implementation of Newton–Cotes Rules in order to reduce the amount of -planes necessary to obtain a good representation of the GT2 FS.
The error behavior tends to decrease when the number of samples is increased, and Fig. 11 shows an illustration of these phenomena. So, the computational cost cannot be reduced by only the reduction of the samples number without an
Experiments
The experiments aim at illustrating the advantages and disadvantages of the proposed approach with respect to the conventional approach. The first experiment consists on comparing both approaches for Type Reduction and the Defuzzification process. The second experiment consists on the comparison of both approaches in a control application and finally, the third experiment consists on a computation time comparison between both approaches.
Conclusion
As conclusions, based on the results obtained in the experiments, the proposed approach for -planes integration shows a significantly improvement with respect to the conventional -planes integration, this means that is possible to model a GT2 FS with a lower number of -planes, reducing in this way the computational cost of the GT2 Fuzzy Inference Systems. This is specially interesting for engineering applications because one of the big limitations of the implementation of GT2 Fuzzy Logic in
References (35)
- et al.
Optimization of fuzzy controller design using a new bee colony algorithm with fuzzy dynamic parameter adaptation
Appl. Soft Comput.
(2016) - et al.
A comparative study of type-1 fuzzy logic systems, interval type-2 fuzzy logic systems and generalized type-2 fuzzy logic systems in control problems
Inform. Sci.
(2016) - et al.
A review on interval type-2 fuzzy logic applications in intelligent control
Inform. Sci.
(2014) - et al.
Type-2 fuzzy logic aggregation of multiple fuzzy controllers for airplane flight control
Inform. Sci.
(2015) - et al.
New geometric inference techniques for type-2 fuzzy sets
Internat. J. Approx. Reason.
(2008) - et al.
Intelligent control for nonlinear inverted pendulum based on interval type-2 fuzzy PD controller
Alex. Eng. J.
(2014) - et al.
Centroid of a type-2 fuzzy set
Inform. Sci.
(2001) - et al.
Adaptive fuzzy logic control of dynamic balance and motion for wheeled inverted pendulums
Fuzzy Sets and Systems
(2009) - et al.
An improved method for edge detection based on interval type-2 fuzzy logic
Expert Syst. Appl.
(2010) - et al.
A knowledge-based system for breast cancer classification using fuzzy logic method
Telemat. Inform.
(2017)