Comparison between a continuous and a discrete method for the aggregation and deffuzification stages of a TRIGA reactor power fuzzy controller

https://doi.org/10.1016/j.pnucene.2005.03.012Get rights and content

Abstract

This paper presents a comparison of the performance of two different methods to realize the aggregation and the centre of gravity stages of a fuzzy controller that is under development for its integration, as an alternative power control algorithm, in the control console of the TRIGA Mark III reactor of the Mexican Nuclear Centre. In one case, an innovative method determines, in every control cycle, the group of lines that define the fuzzy aggregated set of the rule base in the continuous domain of the output variable. Likewise, the centre of gravity of this aggregated set is analytically obtained, and the corresponding controller is named exact or continuous. In the other case, a method is used to determine, in one step, both the aggregated set and its centre of gravity using the classical discretization of the universe of discourse of the output variable, thus leading to the discrete fuzzy controller. These methods were simulated in the ascent and regulation of neutron power in a TRIGA Mark III reactor. The performance parameters used for the comparison between the two methods were: The required number of floating point operations, the time required to attain a certain power level, the neutron power time response, and the reactor period values.

References (14)

  • BubakM. et al.

    A fuzzy-logic approach to HTR nuclear power plant model control

    Annals of Nuclear Energy UK

    (1983)
  • RuanD. et al.

    Controlling the power output of a nuclear reactor with fuzzy logic

    Information Sciences

    (1998)
  • Benitez-ReadJ.S. et al.

    Neutron Power Control in a Research Reactor Using a Fuzzy Rule Based System

    Soft Computing with Industrial Applications

    (1996)
  • Benítez-ReadJ.S. et al.

    Controlling neutron power of a TRIGA Mark III research nuclear reactor with fuzzy adaptation of the set of output membership functions

  • Benítez ReadJ.S. et al.

    Implantación en un DSP de un controlador difuso para la regulación de potencia neutrónica en un reactor tipo TRIGA

    Electro 2002

    (2002)
  • Benítez-ReadJ.S. et al.

    Neutron power control using a fuzzy algorithm with exact aggregation and defuzzification

    Intelligent Automation and Soft Computing

    (2005)
  • BernardJ.A. et al.

    Digital control of power transients in a nuclear reactor

    IEEE Trans. on Nuclear Science

    (1984)
There are more references available in the full text version of this article.

Cited by (8)

  • Union score and fuzzy logic mineral prospectivity mapping using discretized and continuous spatial evidence values

    2017, Journal of African Earth Sciences
    Citation Excerpt :

    In knowledge-driven MPM, as discussed in the Introduction, discretization of continuous spatial data into classes and assign the same weight to all values in each class (e.g., Carranza and Hale, 2001; Luo and Dimitrakopoulos, 2003; Porwal et al., 2003, 2004, 2006; Carranza, 2008b; McKay and Harris, 2015; Elliott et al., 2016; Ford et al., 2016) represents systemic bias in weighting of evidence of prospectivity. However, as have been shown by Tsoukalas and Uhrig (1997), Nykänen et al. (2008), Yousefi and Carranza (2015a), Yousefi and Nykänen (2016), in other fields of study (e.g., Benitez-Read et al., 2005; Xie et al., 2014), and in this paper, assigning continuous weights does not need discretization of continuous evidential values; thus, bias in weighting of evidence due to simplification of data can be avoided. The method for defining continuous weights (or fuzzy membership values) is generally appropriate for continuous-value spatial data.

  • Fuzzification of continuous-value spatial evidence for mineral prospectivity mapping

    2015, Computers and Geosciences
    Citation Excerpt :

    The important advantages of using modified expected value and continuous fuzzy evidence maps to MPM introduced in this paper are as follows. As mentioned in the Introduction, discretization of continuous values in a map to be used as spatial evidence is not needed in fuzzification because fuzzy evidence layers can be generated by using continuous fuzzy membership values as have been shown in Nykänen et al. (2008), here and in other fields of study (e.g., Clenshaw and Olver, 1984; Sakawa and Yauchi, 1999; Benitez-Read et al., 2005; Narmatha Banu and Devaraj, 2012; Ray, 2012; Guillén-Flores et al., 2013; Silva et al., 2014; Xie et al., 2014). However, it has been a traditional practice in MPM to discretize continuous-value spatial evidence into categorized maps (e.g. Bonham-Carter, 1994; Cheng and Agterberg, 1999; D’Ercole et al., 2000; Knox-Robinson, 2000; Carranza and Hale, 2001; Luo and Dimitrakopoulos 2003; Porwal et al., 2003, 2004, 2006; Rogge et al., 2006; Carranza, 2008; González-Álvarez et al., 2010; Markwitz et al., 2010; Lisitsin et al., 2013).

  • Applying the computational intelligence paradigm to nuclear power plant operation: A review (1990-2015)

    2020, Research Anthology on Artificial Intelligence Applications in Security
  • Design and implementation of a fuzzy controller for a TRIGA mark III reactor

    2012, Science and Technology of Nuclear Installations
View all citing articles on Scopus
View full text