Implementing Generic Input Output Functions #1149
Description
Hello,
I am trying to use this library to model alternative implementations of various algorithms for current control of a 3 phase motor.
The basic outline is as follows:
Current-Target -> PI -> Inverse Park/Clarke -> SVM -> Motor Model -> Current Sense Model -> Park/Clarke -> Current Measured (feedback)
Specifically my question is it possible to have essentially generic function blocks that given a set of inputs return a set of outputs independent of time. My hope is to be able to have a few versions of these and drop them into the model for comparison. A trivial example I tried was attempting to get the park transform working. I tried the following of trying to hack a non-linear system to do this but it doesn't work (which I believe is expected):
import matplotlib.pyplot as plt
import numpy as np
import control as ct
def inverse_park_transformation_out(t, x, u, params):
return [x[0],x[1]]
def inverse_park_transformation(t, x, u, params):
q = u[0]
d = u[1]
angle = u[2]
alpha = d * np.cos(angle) - q * np.sin(angle)
beta = d * np.sin(angle) + q * np.cos(angle)
return [alpha, beta]
inv_park = ct.nlsys(updfcn=inverse_park_transformation, outfcn=inverse_park_transformation_out, inputs=[
"q", "d", "angle"], outputs=["alpha", "beta"], states = ["alpha", "beta"],name="inv_park")
dt = 50e-6
f = 100
t = np.arange(0.0, 0.02, dt)
q = np.ones_like(t)
d = np.zeros_like(t)
angle = np.sin(2 * np.pi * f * t) * np.pi
t, y, s = ct.input_output_response(sys=inv_park, T=t, U=[q, d, angle], return_x=True)
plt.plot(t, q, label="V_q")
plt.plot(t, d, label="V_d")
plt.plot(t, angle, label="$\\theta$")
plt.plot(t, y[0], label="v_alpha")
plt.plot(t, y[1], label="v_beta")
plt.legend()
plt.show()