4 replies to this topic

[mariner]

I think this belongs in the tech. article archive as its reather heavy on maths etc. but as I don't know how to do that I will link it here 

 

https://blogs.mathwo.../06/07/mf-tyre/

[desmo]

Done. I hope.

[Greg Locock]

An body with matlab could they try this script? it's for use with MFeval

 

clear

clc

 

close all

% Number of points

nPoints = 2000;

 

% Pure lateral test case

Fz      = ones(nPoints,1).*3000;            % vertical load         (N)

kappa = .5*(rand(1, nPoints)'-.5);               % longitudinal slip (-) (-1 = locked wheel)

alpha = 2*pi*60/360*(rand(1, nPoints)'-.5);     % side slip angle    (radians)

gamma = ones(nPoints,1).*0;               % inclination angle (radians)

phit = ones(nPoints,1).*0;               % turnslip            (1/m)

Vx    = ones(nPoints,1).*16;              % forward velocity    (m/s)

 

% Create a string with the name of the TIR file

TIRfile = 'MagicFormula61_Parameters.TIR';

 

% Select a Use Mode

% useMode specifies the type of calculation performed:

%      1: combined force/moment calculation

%      2: combined force/moment calculation + turn slip

%    +10: revoke alpha_star definition

%    +20: include alpha_star definition

%   +100: include limit checks

%   +200: ignore limit checks

useMode = 122;

 

% Wrap all inputs in one matrix

inputs = [Fz kappa alpha gamma phit Vx];

% Store the output from mfeval in a 2D Matrix

output = mfeval(TIRfile, inputs, useMode);

%Plot results

% Extract variables from output MFeval. For more info type "help mfeval"

Fy = output(:,2);

Fx = output(:,1);

Mz = output(:,6);

Mx = output(:,4);

SA = rad2deg(output(:,8)); % Convert to degrees

t = output(:,16);

 

figure

subplot(2,2,1)

plot(Fx, Fy, '.')

grid on

title('Fx-Fy')

xlabel('traction force')

ylabel('Lateral Force (N)')

 

subplot(2,2,2)

plot(SA, Fy,'.')

grid on

%title('Mz-SA')

xlabel('Slip Angle (deg)')

ylabel('Fy')

 

subplot(2,2,3)

plot(kappa,Fx,'.')

grid on

%title('t-SA')

xlabel('Slip ')

ylabel('Fx')

 

subplot(2,2,4)

plot(SA, kappa,'.')

grid on

%title('Mx-SA')

xlabel('Slip Angle (deg)')

ylabel('slip x')

 

 

It's supposed to fire a bunch of longitudinal slip and steer at the function and see if it produces a believable friction circle. I'm using Octave and get crap results.

Edited by Greg Locock, 15 December 2022 - 10:36.

[TDIMeister]

Greg,

 

mfeval not found...

[Greg Locock]

Sorry, it is a download from the mathworks site. https://au.mathworks...ge/63618-mfeval As it happens we've solved the mystery,te

https://au.mathworks...ge/63618-mfeval

 

as it happens octave and matlab agree, the problem was the monte carlo paid too much emphasis on the post peak behaviour.