Faculty of Engineering: Department of Electrical and Electronic Engineering: Research: Control and Power
About the groupPeopleProjectsPublicationsEventsLinks
Select your text size for this site here: Small Text Normal Text Large Text Extra Large Text

Note: Some of the graphical elements of this site are only visible to browsers that support accepted web standards. The content of this site is, however, accessible to any browser or Internet device.

EPSRC

Control and Power Portfolio Partnership


CONTROL OF DC/THREE-PHASE POWER CONVERTER

PROGRAMME

This project has two main branches running in parallel: a theoretic branch and a power electronics branch.

Theoretic branch

This branch develops multi-periodic repetitive control, which is based on a highly complex internal model capable of rejecting several periodic disturbances of different periods. This internal  model is continuously adjusted by an adaptation subsystem which tracks the relevant frequencies and adjusts certain weights according to the amplidudes of the different periodic components in the reference and  disturbance signals. This part of the research uses techniques from four areas:
  • distributed parameter systems
  • repetitive control
  • robust control (H-infinity control)
  • phase-locked loops.

    • The objectives of this branch are as follows:

    1. Allow the external signals to be multi-periodic. This crucial first step has been largely achieved, but many details have yet to be worked out. The idea is that if in the repetitive feedback system we take M to be a convex combination of internal models with different , then the stability  and w-stability conditions for the system remain the same as for a single period. Good estimates for the size (i.e., power) of the steady state error should be worked out. Another problem is to obtain useful estimates on the H-infinity norm of the closed-loop transfer function from w to e.
    2. Allow the periods to be slowly changing and design a system which will track the periods. This might be an array of PLL circuits with a mechanism to prevent two or more PLLs from locking onto the same frequency.
    3. Work out a method for tuning the weights in the convex combination of internal models, and the filters in the individual internal models. This would probably need some adaptive signal processing techniques, using power spectral densities.

    Power electronics branch

    This branch develops software tools for the design and testing of repetitive controllers for power converters and for active power filters. We would also build a power converter for a small generator and test it.

    The objectives of this branch are as follows:

    1. Develop a library of models of various prime movers of embedded generators, of power converters, network components (e.g. capacitance-dominated and inductance-dominated lines, transformers) and of a variety of loads (e.g. controlled and un-controlled motors, rectifiers, lighting ballasts). The models will be used to test controllers in a variety of circumstances and assess robustness on unknown networks.
    2. Develop design procedures and the corresponding software for the DSP-based control of Pulse Width Modulation DC to AC power converters, working either as part of an isolated generator or supplying power into the network. Develop similar design procedures for active power filters.
    3. Design and implement a DC to three phase power converter to work in conjunction with a 20 kW generator to supply a 50Hz output voltage. This voltage should have a total harmonic distortion of less than 2% for any periodic load current with harmonics up to 2000Hz. It should also cope with a subharmonic load and with DC supply voltage ripple. Small-scale tests and simulations at the university will be followed by testing on a real generator at the TGC test facility.

    [up]