-
Notifications
You must be signed in to change notification settings - Fork 3
/
00_RLC_generate_train.py
104 lines (85 loc) · 2.91 KB
/
00_RLC_generate_train.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
from scipy.integrate import solve_ivp
from scipy.interpolate import interp1d
import numpy as np
import matplotlib.pyplot as plt
import control.matlab
import pandas as pd
import os
from examples.RLC.symbolic_RLC import fxu_ODE, fxu_ODE_nl
if __name__ == '__main__':
# Set seed for reproducibility
np.random.seed(0)
# Input characteristics #
len_sim = 5e-3
Ts = 1e-6 # 5e-7
omega_input = 80e3
std_input = 80
scale_u = 80.
scale_x = [90., 3.]
tau_input = 1 / omega_input
Hu = control.TransferFunction([1], [1 / omega_input, 1])
Hu = Hu * Hu
Hud = control.matlab.c2d(Hu, Ts)
N_sim = int(len_sim // Ts)
N_skip = int(20 * tau_input // Ts) # skip initial samples to get a regime sample of d
N_sim_u = N_sim + N_skip
e = np.random.randn(N_sim_u)
te = np.arange(N_sim_u) * Ts
_, u = control.forced_response(Hu, te, e, return_x=False)
u = u[N_skip:]
u = u / np.std(u) * std_input
t_sim = np.arange(N_sim) * Ts
u_func = interp1d(t_sim, u, kind='zero', fill_value="extrapolate")
def f_ODE(t, x):
u = u_func(t).ravel()
return fxu_ODE(t, x, u)
def f_ODE_nl(t, x):
u = u_func(t).ravel()
return fxu_ODE_nl(t, x, u)
x0 = np.zeros(2)
f_ODE(0.0, x0)
t_span = (t_sim[0], t_sim[-1])
y1 = solve_ivp(f_ODE, t_span, x0, t_eval=t_sim) # Linear
y2 = solve_ivp(f_ODE_nl, t_span, x0, t_eval=t_sim) # Non-Linear
x1 = y1.y.T # Linear
x2 = y2.y.T # Non-Linear
x1 = x1 / scale_x # Linear
x2 = x2 / scale_x # Non-Linear
u = u / scale_u
# In[plot]
# System State x = [v_c, i_l].T
fig, ax = plt.subplots(3, 1, figsize=(10, 10), sharex=True)
ax[0].plot(t_sim, x1[:, 0], 'b') # Linear
ax[0].plot(t_sim, x2[:, 0], 'r') # Non-Linear
ax[0].set_xlabel('time (s)')
ax[0].set_ylabel('Capacitor voltage (V)')
ax[1].plot(t_sim, x1[:, 1], 'b')
ax[1].plot(t_sim, x2[:, 1], 'r')
ax[1].set_xlabel('time (s)')
ax[1].set_ylabel('Inductor current (A)')
ax[2].plot(t_sim, u, 'b')
ax[2].set_xlabel('time (s)')
ax[2].set_ylabel('Input voltage (V)')
ax[0].grid(True)
ax[1].grid(True)
ax[2].grid(True)
plt.show()
# In[Save]
if not os.path.exists("data"):
os.makedirs("data")
X = np.hstack((t_sim.reshape(-1, 1), x1, u.reshape(-1, 1), x1[:, 0].reshape(-1, 1)))
COL_T = ['time']
COL_X = ['V_C', 'I_L']
COL_U = ['V_IN']
COL_Y = ['V_C']
COL = COL_T + COL_X + COL_U + COL_Y
df_X = pd.DataFrame(X, columns=COL)
df_X.to_csv(os.path.join("data", "RLC_data_train_lin.csv"), index=False)
X = np.hstack((t_sim.reshape(-1, 1), x2, u.reshape(-1, 1), x2[:, 0].reshape(-1, 1)))
COL_T = ['time']
COL_X = ['V_C', 'I_L']
COL_U = ['V_IN']
COL_Y = ['V_C']
COL = COL_T + COL_X + COL_U + COL_Y
df_X = pd.DataFrame(X, columns=COL)
df_X.to_csv(os.path.join("data", "RLC_data_train_nl.csv"), index=False)