2018-05-30 14:36:06 +02:00
|
|
|
/****************************************************************************
|
|
|
|
* control/lib_observer.c
|
|
|
|
*
|
|
|
|
* Copyright (C) 2018 Gregory Nutt. All rights reserved.
|
|
|
|
* Author: Mateusz Szafoni <raiden00@railab.me>
|
|
|
|
*
|
|
|
|
* Redistribution and use in source and binary forms, with or without
|
|
|
|
* modification, are permitted provided that the following conditions
|
|
|
|
* are met:
|
|
|
|
*
|
|
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
|
|
* notice, this list of conditions and the following disclaimer.
|
|
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
|
|
* notice, this list of conditions and the following disclaimer in
|
|
|
|
* the documentation and/or other materials provided with the
|
|
|
|
* distribution.
|
|
|
|
* 3. Neither the name NuttX nor the names of its contributors may be
|
|
|
|
* used to endorse or promote products derived from this software
|
|
|
|
* without specific prior written permission.
|
|
|
|
*
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
|
|
|
|
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
|
|
|
|
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
|
|
|
|
* FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
|
|
|
|
* COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
|
|
|
|
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
|
|
|
|
* BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS
|
|
|
|
* OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED
|
|
|
|
* AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
|
|
|
|
* ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
|
|
|
|
* POSSIBILITY OF SUCH DAMAGE.
|
|
|
|
*
|
|
|
|
****************************************************************************/
|
|
|
|
|
|
|
|
/****************************************************************************
|
|
|
|
* Included Files
|
|
|
|
****************************************************************************/
|
|
|
|
|
|
|
|
#include <stddef.h>
|
|
|
|
#include <assert.h>
|
|
|
|
|
|
|
|
#include <dsp.h>
|
|
|
|
|
|
|
|
/****************************************************************************
|
|
|
|
* Pre-processor Definitions
|
|
|
|
****************************************************************************/
|
|
|
|
|
|
|
|
/****************************************************************************
|
|
|
|
* Private Functions
|
|
|
|
****************************************************************************/
|
|
|
|
|
|
|
|
/****************************************************************************
|
|
|
|
* Public Functions
|
|
|
|
****************************************************************************/
|
|
|
|
|
|
|
|
/****************************************************************************
|
|
|
|
* Name: motor_observer_init
|
|
|
|
*
|
|
|
|
* Description:
|
|
|
|
* Initialize motor observer
|
|
|
|
*
|
|
|
|
* Input Parameters:
|
|
|
|
* observer - pointer to the common observer data
|
|
|
|
* ao - pointer to the angle specific observer data
|
|
|
|
* so - pointer to the speed specific observer data
|
|
|
|
* per - observer execution period
|
|
|
|
*
|
|
|
|
* Returned Value:
|
|
|
|
* None
|
|
|
|
*
|
|
|
|
****************************************************************************/
|
|
|
|
|
|
|
|
void motor_observer_init(FAR struct motor_observer_s *observer,
|
|
|
|
FAR void *ao, FAR void *so, float per)
|
|
|
|
{
|
|
|
|
DEBUGASSERT(observer != NULL);
|
|
|
|
DEBUGASSERT(ao != NULL);
|
|
|
|
DEBUGASSERT(so != NULL);
|
|
|
|
DEBUGASSERT(per > 0.0);
|
|
|
|
|
|
|
|
/* Set observer period */
|
|
|
|
|
|
|
|
observer->per = per;
|
|
|
|
|
|
|
|
/* Connect angle estimation observer data */
|
|
|
|
|
|
|
|
observer->ao = ao;
|
|
|
|
|
|
|
|
/* Connect speed estimation observer data */
|
|
|
|
|
|
|
|
observer->so = so;
|
|
|
|
}
|
|
|
|
|
|
|
|
/****************************************************************************
|
|
|
|
* Name: motor_observer_smo_init
|
|
|
|
*
|
|
|
|
* Description:
|
|
|
|
* Initialize motor sliding mode observer.
|
|
|
|
*
|
|
|
|
* Input Parameters:
|
|
|
|
* smo - pointer to the sliding mode observer private data
|
|
|
|
* kslide - SMO gain
|
|
|
|
* err_max - linear region upper limit
|
|
|
|
*
|
|
|
|
* Returned Value:
|
|
|
|
* None
|
|
|
|
*
|
|
|
|
****************************************************************************/
|
|
|
|
|
|
|
|
void motor_observer_smo_init(FAR struct motor_observer_smo_s *smo,
|
|
|
|
float kslide,
|
|
|
|
float err_max)
|
|
|
|
{
|
|
|
|
DEBUGASSERT(smo != NULL);
|
|
|
|
DEBUGASSERT(kslide > 0.0);
|
|
|
|
DEBUGASSERT(err_max > 0.0);
|
|
|
|
|
|
|
|
/* Initialize structure */
|
|
|
|
|
|
|
|
smo->k_slide = kslide;
|
|
|
|
smo->err_max = err_max;
|
|
|
|
}
|
|
|
|
|
|
|
|
/****************************************************************************
|
|
|
|
* Name: motor_observer_smo
|
|
|
|
*
|
|
|
|
* Description:
|
|
|
|
* One step of the SMO observer.
|
|
|
|
* REFERENCE: http://ww1.microchip.com/downloads/en/AppNotes/01078B.pdf
|
|
|
|
*
|
|
|
|
* Below some theoretical backgrounds about SMO.
|
|
|
|
*
|
|
|
|
* The digitalized motor model can be represent as:
|
|
|
|
*
|
|
|
|
* d(i_s.)/dt = (-R/L)*i_s. + (1/L)*(v_s - e_s. - z)
|
|
|
|
*
|
|
|
|
* We compare estimated current (i_s.) with measured current (i_s):
|
|
|
|
*
|
|
|
|
* err = i_s. - i_s
|
|
|
|
*
|
|
|
|
* and get correction factor (z):
|
|
|
|
*
|
|
|
|
* sign = sing(err)
|
|
|
|
* z = sign*K_SLIDE
|
|
|
|
*
|
|
|
|
* Once the digitalized model is compensated, we estimate BEMF (e_s.) by
|
|
|
|
* filtering z:
|
|
|
|
*
|
|
|
|
* e_s. = low_pass(z)
|
|
|
|
*
|
|
|
|
* The estimated BEMF is filtered once again and used to approximate the
|
|
|
|
* motor angle:
|
|
|
|
*
|
|
|
|
* e_filtered_s. = low_pass(e_s.)
|
|
|
|
* theta = arctan(-e_alpha/e_beta)
|
|
|
|
*
|
|
|
|
* The estimated theta is phase-shifted due to low pass filtration, so we
|
|
|
|
* need some phase compensation. More details below.
|
|
|
|
*
|
|
|
|
* where:
|
|
|
|
* v_s - phase input voltage vector
|
|
|
|
* i_s. - estimated phase current vector
|
|
|
|
* i_s - phase current vector
|
|
|
|
* e_s. - estimated phase BEMF vector
|
|
|
|
* R - motor winding resistance
|
|
|
|
* L - motor winding inductance
|
|
|
|
* z - output correction factor voltage
|
|
|
|
*
|
|
|
|
* Input Parameters:
|
|
|
|
* observer - (in/out) pointer to the common observer data
|
|
|
|
* i_ab - (in) inverter alpha-beta current
|
|
|
|
* v_ab - (in) inverter alpha-beta voltage
|
|
|
|
* phy - (in) pointer to the motor physical parameters
|
|
|
|
*
|
|
|
|
* Returned Value:
|
|
|
|
* None
|
|
|
|
*
|
|
|
|
****************************************************************************/
|
|
|
|
|
|
|
|
void motor_observer_smo(FAR struct motor_observer_s *observer, FAR ab_frame_t *i_ab,
|
|
|
|
FAR ab_frame_t *v_ab, FAR struct motor_phy_params_s *phy)
|
|
|
|
{
|
|
|
|
FAR struct motor_observer_smo_s *smo =
|
|
|
|
(FAR struct motor_observer_smo_s *)observer->ao;
|
|
|
|
FAR ab_frame_t *emf = &smo->emf;
|
|
|
|
FAR ab_frame_t *z = &smo->z;
|
|
|
|
FAR ab_frame_t *i_est = &smo->i_est;
|
|
|
|
FAR ab_frame_t *v_err = &smo->v_err;
|
|
|
|
FAR ab_frame_t *i_err = &smo->i_err;
|
|
|
|
FAR ab_frame_t *sign = &smo->sign;
|
|
|
|
float i_err_a_abs = 0.0;
|
|
|
|
float i_err_b_abs = 0.0;
|
|
|
|
float angle = 0.0;
|
|
|
|
|
|
|
|
/* REVISIT: observer works only when IQ current is high enough */
|
|
|
|
|
|
|
|
/* Calculate observer gains */
|
|
|
|
|
|
|
|
smo->F_gain = (1 - observer->per*phy->res/phy->ind);
|
|
|
|
smo->G_gain = observer->per/phy->ind;
|
|
|
|
|
|
|
|
/* Saturate F gain */
|
|
|
|
|
|
|
|
if (smo->F_gain < 0)
|
|
|
|
{
|
|
|
|
smo->F_gain = 0.0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Saturate G gain */
|
|
|
|
|
|
|
|
if (smo->G_gain > 0.999)
|
|
|
|
{
|
|
|
|
smo->G_gain = 0.999;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Configure low pass filters
|
|
|
|
*
|
|
|
|
* We tune low-pass filters to achieve cutoff frequency equal to
|
|
|
|
* input singal frequency. This gives us constant phase shift between
|
|
|
|
* input and outpu signals equals to:
|
|
|
|
*
|
|
|
|
* phi = -arctan(f_in/f_c) = -arctan(1) = -PI/2
|
|
|
|
*
|
|
|
|
* Input signal frequency is equal to the frequency of the motor currents,
|
|
|
|
* which give us:
|
|
|
|
*
|
|
|
|
* f_c = omega_e/(2*PI)
|
|
|
|
* omega_m = omega_e/poles
|
|
|
|
* f_c = omega_m*poles/(2*PI)
|
|
|
|
*
|
|
|
|
* filter = T * (2*PI) * f_c
|
|
|
|
* filter = T * omega_m * poles
|
|
|
|
*
|
|
|
|
* T - [s] period at which the digital filter is being calculated
|
|
|
|
* f_in - [Hz] input frequency of the filter
|
|
|
|
* f_c - [Hz] cutoff frequency of the filter
|
|
|
|
* omega_m - [rad/s] mechanical angular velocity
|
|
|
|
* omega_e - [rad/s] electrical angular velocity
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
|
|
|
|
smo->emf_lp_filter1 = observer->per * observer->speed * phy->poles;
|
|
|
|
smo->emf_lp_filter2 = smo->emf_lp_filter1;
|
|
|
|
|
|
|
|
/* Get voltage error: v_err = v_ab - emf */
|
|
|
|
|
|
|
|
v_err->a = v_ab->a - emf->a;
|
|
|
|
v_err->b = v_ab->b - emf->b;
|
|
|
|
|
|
|
|
/* Estimate stator current */
|
|
|
|
|
|
|
|
i_est->a = smo->F_gain * i_est->a + smo->G_gain * (v_err->a - z->a);
|
|
|
|
i_est->b = smo->F_gain * i_est->b + smo->G_gain * (v_err->b - z->b);
|
|
|
|
|
|
|
|
/* Get motor current errror */
|
|
|
|
|
|
|
|
i_err->a = i_ab->a - i_est->a;
|
|
|
|
i_err->b = i_ab->b - i_est->b;
|
|
|
|
|
|
|
|
/* Slide-mode controller */
|
|
|
|
|
|
|
|
sign->a = (i_err->a > 0 ? 1.0 : -1.0);
|
|
|
|
sign->b = (i_err->b > 0 ? 1.0 : -1.0);
|
|
|
|
|
|
|
|
/* Get current error absolute value - just multiply value with its sign */
|
|
|
|
|
|
|
|
i_err_a_abs = i_err->a * sign->a;
|
|
|
|
i_err_b_abs = i_err->b * sign->b;
|
|
|
|
|
|
|
|
/* Calculate new output correction factor voltage */
|
|
|
|
|
|
|
|
if (i_err_a_abs < smo->err_max)
|
|
|
|
{
|
|
|
|
/* Enter linear region if error is small enough */
|
|
|
|
|
|
|
|
z->a = i_err->a * smo->k_slide / smo->err_max;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
/* Non-linear region */
|
|
|
|
|
|
|
|
z->a = sign->a * smo->k_slide;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (i_err_b_abs < smo->err_max)
|
|
|
|
{
|
|
|
|
/* Enter linear region if error is small enough */
|
|
|
|
|
|
|
|
z->b = i_err->b * smo->k_slide / smo->err_max;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
/* Non-linear region */
|
|
|
|
|
|
|
|
z->b = sign->b * smo->k_slide;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Filter z to obtain estimated emf */
|
|
|
|
|
|
|
|
LP_FILTER(emf->a, z->a, smo->emf_lp_filter1);
|
|
|
|
LP_FILTER(emf->b, z->b, smo->emf_lp_filter1);
|
|
|
|
|
|
|
|
/* Filter emf one more time before angle stimation */
|
|
|
|
|
|
|
|
LP_FILTER(emf->a, emf->a, smo->emf_lp_filter2);
|
|
|
|
LP_FILTER(emf->b, emf->b, smo->emf_lp_filter2);
|
|
|
|
|
|
|
|
/* Estimate phase angle according to:
|
|
|
|
* emf_a = -|emf| * sin(th)
|
|
|
|
* emf_b = |emf| * cos(th)
|
|
|
|
* th = atan2(-emf_a, emf->b)
|
|
|
|
*/
|
|
|
|
|
|
|
|
angle = fast_atan2(-emf->a, emf->b);
|
|
|
|
|
|
|
|
#if 1
|
|
|
|
/* Some assertions
|
2018-05-30 14:49:48 +02:00
|
|
|
* TODO: simplify
|
2018-05-30 14:36:06 +02:00
|
|
|
*/
|
|
|
|
|
|
|
|
if (angle != angle) angle = 0.0;
|
|
|
|
if (emf->a != emf->a) emf->a = 0.0;
|
|
|
|
if (emf->b != emf->b) emf->b = 0.0;
|
|
|
|
if (z->a != z->a) z->a = 0.0;
|
|
|
|
if (z->b != z->b) z->b = 0.0;
|
|
|
|
if (i_est->a != i_est->a) i_est->a = 0.0;
|
|
|
|
if (i_est->b != i_est->b) i_est->b = 0.0;
|
|
|
|
#endif
|
|
|
|
|
|
|
|
/* Angle compensation.
|
|
|
|
* Due to low pass filtering we have some delay in estimated phase angle.
|
|
|
|
*
|
|
|
|
* Adaptive filters introduced above cause -PI/2 phase shift for each filter.
|
|
|
|
* We use 2 times filtering which give us constant -PI phase shift.
|
|
|
|
*/
|
|
|
|
|
|
|
|
angle = angle -M_PI_F;
|
|
|
|
|
|
|
|
/* Normalize angle to range <0, 2PI> */
|
|
|
|
|
|
|
|
angle_norm_2pi(&angle, 0.0, 2*M_PI_F);
|
|
|
|
|
|
|
|
/* Store estimated angle in observer data*/
|
|
|
|
|
|
|
|
observer->angle = angle;
|
|
|
|
}
|