Project Logs

Collapse

• DC supply voltage control

  Mitja Breznik • 05/12/2019 at 12:44 • 0 comments

  I have decided to first develop a simple DC supply voltage controller that will be the basis for two configurations described in the introduction (controller acting as a voltage source for the self sustained DC "grid"; controller connected to the grid-tied inverter).

  Controller is a simple PI regulator that determines the motor torque based on DC voltage reference and measured DC voltage:

  Parameters are currently set with experimentation, but there should be a simple way of defining the coefficients based on system parameters (capacitance, torque constant, rotational speed).
  

  For the "proof of concept" test I used a simple power resistor as a load for DC supply:

  Voltage reference was set at 15 V (voltage controller running at 1kHz), while resistor was connected and disconnected multiple times. Voltage remained constant while the motor current changed according the the DC supply energy demand:

  Here is a video of the multimeter  and power resistor connected to the controller DC terminals:

  After some additional testing I will add the support of the new mode and parameters to the PC GUI.
  

• MPPT scanning algorithm test

  Mitja Breznik • 05/05/2019 at 09:24 • 0 comments

  Simplest way for me to test different energy conversion functionalities is to build a back-to-back setup. I have coupled two brushless outrunners enclosed in a plexiglass tube for protection. It is not the optimum solution for long term testing, due to heat build-up, but it will suffice for SW testing.

  Instead of connecting the generator side of the system to the wind or water turbine I programmed a simple linear speed/torque curve to one of the motors:
  
  set_speed = no_load_speed - ramp_coefficient*motor_current
  
  Parameters of the load curve can be adjusted with a PC GUI. Linear load curve should have a maximum power point at no_load_speed/2. Wind or water turbines tend to have higher order load curves but this simple curve is sufficient for first tests.
  
  I have programmed a simple MPPT algorithm, that measures the generating power over the whole pre-programmed range and then settles at the MPP:
  All parameters can be set of the PC GUI application with the modified VESC Tool:
  Number of different parameters should ensure that SW could be adapted to a number of different generator applications.
  
  Here is a screenshot of the motor currents/speed during the first MPPT test:
  
  In order to use this algorithm to charge the battery there is feature added which gradually reduces the charging current until reaching full battery voltage.
  
  I will perform the next MPPT test by connecting the generator side to the LiPo battery pack and test the full battery derating function.
  

View all 2 project logs