Investigation of Membership Function Shapes in a Fuzzy-Controlled HVDC System

Abstract

The use of fuzzy logic controllers to improve the performance of HVDC systems under various fault and operating point changes has been well-documented. However, previous research has focused either on using only basic triangular membership functions, or on optimizing the membership functions for a particular event. This paper investigates the relationship between various different shapes of membership functions and the associated HVDC system response to different events. It is concluded that a membership function optimized for a certain event will likely not perform well under various different conditions. Other trends are also observed and discussed to provide useful recommendations for fuzzy HVDC control systems.

Fuzzy logic (FL) controlled HVDC system-influence of shape, width & distribution of membership functions (MFs)

“…The Larsen Inference Engine is used first as it is simple and powerful among the various other implication operators [9]. It is given as (9) Where, e, are the error and rate of change of error inputs respectively, and u is the output.…”

“…It is given as (9) Where, e, are the error and rate of change of error inputs respectively, and u is the output. represents 10) is also employed.…”

“…Hence, there is a need to study the effect of different MF shapes, width and distribution on the performance of a HVDC system. In [9], the authors compare Gaussian and triangular MFs of different shapes and widths and conclude that width of the MF matters, and for different faults and operating point changes there would be a different best performing MF, but this work does not talk about some other available MFs shapes and most importantly distribution of MFs.…”

Influence Of Membership Function’s Shape On Portfolio Optimization Results

“…Research on the influence of membership function shapes is discussed in the area of fuzzy controls. Koprinkova [10], Marshall [11], Multani [12], among others, considered it from different points of view. The first two works conclude that nonlinear MFs i.e.…”

Dc Microgrid Stabilization Through Fuzzy Control of Interleaved, Heterogeneous Storage Elements

“…The degree of membership a value can have is in the range [0,1], where 0 denotes no "belonging" at all, middle values indicate partial "belonging," and 1 denotes a complete A variety of shapes can be used to specify membership functions. Marshall, Kazerani, and Shatshat [35] performed an investigation into varying the shapes of membership functions to achieve certain optimizations in a HVDC system controlled via fuzzy logic. Their conclusions were that it was exceedingly unlikely for a single type of membership function shape to work well for every situation, and that the choice of membership functions should be chosen based on the primary system requirements.…”