Learning neural network for fuzzy controller with three outputs

There is a problem to create a fuzzy controller for controlling the coefficients of the PID controller. I.e. have three input variables (error control, integral and derivative of error) and three output (proportional to the coefficient. PID, integral, differential). There is a corresponding teaching signal, but by anfis fuzzy controller does not do it, because neural network module displays all the inputs in the only output. How can I solve this problem?

Answer 1

[Dimitris]

ANFIS is built to have a single output. Use a standard neural network if you need 3 outputs.