hello Federico
this is my suggestion : I give a given RPM target value and serach for all points in the 2D array that are close to this value (with a given tolerance); then I can plot the same coordinate points from the torque and efficiency curves
the second figure I added a bit of smoothing
example for RPM = 3000
rpm_tol = 1;
rpm_target = 3000;
[row, column] = find(abs(rpm-rpm_target) < rpm_tol);
rpm_values = rpm(sub2ind(size(rpm), row, column));
C_mot_values = C_mot(sub2ind(size(C_mot), row, column));
eta_mech_values = eta_mech(sub2ind(size(eta_mech), row, column));
figure,
subplot(311),plot(rpm_values);
ylabel('RPM');
subplot(312),plot(C_mot_values);
ylabel('Torque');
subplot(313),plot(eta_mech_values);
ylabel('Efficiency');
figure,
meth = 'gaussian';
N = 10;
subplot(311),plot(smoothdata(rpm_values,meth,N));
ylabel('RPM');
subplot(312),plot(smoothdata(C_mot_values,meth,N));
ylabel('Torque');
subplot(313),plot(smoothdata(eta_mech_values,meth,N));
ylabel('Efficiency');