Generating rules in FIS

2 次查看（过去 30 天）
NETHRAVATHI S
NETHRAVATHI S 2021-4-22
评论： Sam Chak 2024-9-20
Hello,
Is there a method to generate rules for FIS.
I have 6 inputs with 5 MFs each. It is confusing to create rules.
So basically i will have 6^5 rules which is huge.
Can we generate rules using some program/logic.
  1 个评论
Sam Chak
Sam Chak 2024-9-20
Ran in:
The total number of possible rules under "Grid Partitioning" technique is
5^6
ans = 15625
NOT
6^5
ans = 7776

请先登录，再进行评论。

请先登录，再回答此问题。

回答（1 个）

Rushikesh
Rushikesh 2024-9-19
编辑：Rushikesh 2024-9-19
Hello @NETHRAVATHI S,
As per my understanding, you want to create rules for fuzzy system automatically and want to reduce complexity involved in creating large number of rules.
Assuming that you have input-output data representing the system you want to model, MATLAB provides the "genfis" function to generate a Fuzzy Inference System (FIS) structure. This function creates rules based on all possible combinations of input membership functions. You can specify membership functions for each input using the options parameter in the "genfis" function. By default, it uses Grid Partitioning to generate the fuzzy system, but it also supports other clustering methods, such as the Subtractive Clustering method, which may help in reducing the number of rules.
You can refer to the documentation below for more information:
https://www.mathworks.com/help/fuzzy/genfis.html
Example of Grid Partitioning method:-
options = genfisOptions('GridPartition');
options.NumMembershipFunctions = [3 3]; % Example: 3 MFs for each of 2 inputs
options.InputMembershipFunctionType = 'gaussmf'; % Example: Gaussian MFs
% Generate initial FIS
initialFIS = genfis(inputData, outputData, options);
Alternatively, if you already have a collection of input-output data that you would like to use for modeling, you can also try using the "anfis" function. This function will automatically generate and optimize rules for a Fuzzy Inference System based on input-output training data using neuro-adaptive learning techniques.
For more information, please refer to the documentation below:
https://www.mathworks.com/help/fuzzy/neuro-adaptive-learning-and-anfis.html
Hope this helps

类别

在 Help Center 和 File Exchange 中查找有关 Fuzzy Logic Toolbox 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by