Proceedings of the 2015 Conference of the International Fuzzy Systems Association and the European Society for Fuzzy Logic and 2015
DOI: 10.2991/ifsa-eusflat-15.2015.128
|View full text |Cite
|
Sign up to set email alerts
|

Memetic Type-2 Fuzzy System Learning for Load Forecasting

Abstract: This paper presents an automatic method to design interval type-2 fuzzy systems for load forecasting applications using a memetic algorithm. This hybridisation of a variable-length genetic algorithm and a gradient descent method allows for concurrent learning of the system's parameters and structure in a versatile fashion. Results are presented addressing chaotic system and market-level one-day-ahead load forecasting.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2

Citation Types

0
2
0

Year Published

2016
2016
2019
2019

Publication Types

Select...
2
1

Relationship

0
3

Authors

Journals

citations
Cited by 3 publications
(2 citation statements)
references
References 20 publications
0
2
0
Order By: Relevance
“…Note that the MA_HT2BFS is compared with different Type-1 FLSs, Type-2 FLSs and neural network learning approaches. For Type-2 FLSs, our system is principally compared with the SA-IT2FLS [53] which is an interval type-2 fuzzy system optimized by the simulated annealing algorithm, and with the memetic-T2FS [54] which uses a variable-length genetic algorithm with a gradient descent technique for the structure and parameters learning of the interval type-2 fuzzy system. Our system is also compared to the support vector-based interval type-2 fuzzy system: TSK-SVR II [55] and to a general type-2 fuzzy system that uses vertical-slices centroid type-reduction method: GT2FLS-VSCTR [56].…”
Section: A Mackey-glass Chaotic Time Series 1) Case 1: Noise-free Mamentioning
confidence: 99%
See 1 more Smart Citation
“…Note that the MA_HT2BFS is compared with different Type-1 FLSs, Type-2 FLSs and neural network learning approaches. For Type-2 FLSs, our system is principally compared with the SA-IT2FLS [53] which is an interval type-2 fuzzy system optimized by the simulated annealing algorithm, and with the memetic-T2FS [54] which uses a variable-length genetic algorithm with a gradient descent technique for the structure and parameters learning of the interval type-2 fuzzy system. Our system is also compared to the support vector-based interval type-2 fuzzy system: TSK-SVR II [55] and to a general type-2 fuzzy system that uses vertical-slices centroid type-reduction method: GT2FLS-VSCTR [56].…”
Section: A Mackey-glass Chaotic Time Series 1) Case 1: Noise-free Mamentioning
confidence: 99%
“…This is due to the hierarchical nature of the system and the use of a multi-objective optimization process which has a great impact on the reduction of the resulting rule base without affecting the system's prediction performance. 1.7e-02 LNF [62] 7.0e-04 7.9e-04 NARMA [63] 6.3e-04 6.2e-04 FBBFNT [57] 9.9e-07 2.0e-06 MA_EFBBFNT [58] 4.1e-11 4.1e-11 GT2FLS-VSCTR [56] 3.9e-02 3.9e-02 TSK-SVR II [55] -7.0e-03 Memetic-T2FS [54] 3.1e-03 -SA-IT2FLS [53] 9.0e-03 8.9e-03 MA_HT2BFS 6.9e-16 6.7 e-16 Other comparisons with existing neural network learning approaches are also discussed, and we can remark from Table I that the evolutionary neural system FBBFNT [57] can generate high rates of accuracy but using a huge number of NFEs (more than 800,000). In the case of the MA_EFBBFNT [58] which is an extended version of FBBFNT that integrates a multi-agent architecture for training, this system succeeds to generate similar low training and testing errors and with much fewer number of function evaluations.…”
Section: A Mackey-glass Chaotic Time Series 1) Case 1: Noise-free Mamentioning
confidence: 99%