How to prepare data from excel for fuzzy analysis.

I have an problem with preparing data for fuzzy analysis. I have 16 groups of data which are inputs and 2 groups of data which are outputs. It is data which was measured and I have abou 3500 lines. I have these data in MS excel and I have to prepare it for FUZZY. Do you know how to do it, please? I have no experience with matlab and fuzzy analysis, but I need it for my master thesis, it will be really nice if somebody help me.

[Petra]

The problem is, that I have to find parameter, which has the biggest influence to outputs. I have two outputs, they are interdependent.

Answer 1

[Prateekshya]

Hello Petra,

Preparing data for fuzzy analysis involves organizing your data in a way that can be used to create a fuzzy inference system (FIS). Since you are using MATLAB, you'll likely be working with the Fuzzy Logic Toolbox, which provides tools to create and evaluate fuzzy systems. Here's a step-by-step guide to help you prepare your data and set up a fuzzy system:

Step 1: Import Data from Excel

First, you will need to import your data from Excel into MATLAB. You can do this using the `readtable` function, which reads data into a table format.

% Import data from Excel
data = readtable('your_data_file.xlsx');

Step 2: Organize Your Data

Assuming your Excel file has headers, you can access your input and output data:

% Extract input and output data
inputs = data{:, 1:16};  % Assuming the first 16 columns are inputs
outputs = data{:, 17:18}; % Assuming the next 2 columns are outputs

Step 3: Create a Fuzzy Inference System

You can create a fuzzy inference system using the Fuzzy Logic Designer in MATLAB or programmatically. Here, I'll show you how to do it programmatically.

% Create a new FIS
fis = newfis('MyFuzzySystem');

% Add input variables
for i = 1:16
    fis = addvar(fis, 'input', ['Input' num2str(i)], [min(inputs(:,i)) max(inputs(:,i))]);
    % Add membership functions for each input
    fis = addmf(fis, 'input', i, 'Low', 'trimf', [min(inputs(:,i)) min(inputs(:,i)) mean(inputs(:,i))]);
    fis = addmf(fis, 'input', i, 'Medium', 'trimf', [min(inputs(:,i)) mean(inputs(:,i)) max(inputs(:,i))]);
    fis = addmf(fis, 'input', i, 'High', 'trimf', [mean(inputs(:,i)) max(inputs(:,i)) max(inputs(:,i))]);
end

% Add output variables
for j = 1:2
    fis = addvar(fis, 'output', ['Output' num2str(j)], [min(outputs(:,j)) max(outputs(:,j))]);
    % Add membership functions for each output
    fis = addmf(fis, 'output', j, 'Low', 'trimf', [min(outputs(:,j)) min(outputs(:,j)) mean(outputs(:,j))]);
    fis = addmf(fis, 'output', j, 'Medium', 'trimf', [min(outputs(:,j)) mean(outputs(:,j)) max(outputs(:,j))]);
    fis = addmf(fis, 'output', j, 'High', 'trimf', [mean(outputs(:,j)) max(outputs(:,j)) max(outputs(:,j))]);
end

Step 4: Define Fuzzy Rules

Fuzzy rules define how the inputs are related to the outputs. You can add rules like this:

% Add rules
rules = [1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1; % Example rule
    0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 1 1 1]; % Another example rule
fis = addrule(fis, rules);

Step 5: Evaluate the FIS

Once your FIS is set up, you can evaluate it with your data.

% Evaluate the fuzzy system
outputResults = evalfis(inputs, fis);

Step 6: Analyze and Visualize

You can visualize the results and analyze the performance of your fuzzy system using various MATLAB plotting functions.

% Plot results
figure;
plot(outputs, 'o'); % Actual outputs
hold on;
plot(outputResults, 'x'); % FIS outputs
legend('Actual', 'FIS Predicted');
title('Fuzzy System Output vs. Actual Output');

I hope this helps!