Integration of model-predictive control in medium and high-voltage power electronics products: An industrial perspective on gaps and progress required
- ,
- Georgios D. Demetriades,
- Vassilios G. Agelidis
- ABB Group,
- ABB Corporate Research,
- University of New South Wales
Information
Output type
Original language
EnglishArticle number
7392675Pages from-to (Number of pages)
Pages 3689-3695 (7 pages)Publication milestones
- Published - 28/01/2016
Publication status
Publisher
Institute of Electrical and Electronics Engineers, United StatesISBN (Electronic)
9781479917624Host publication title
IECON 2015 - 41st Annual Conference of the IEEE Industrial Electronics SocietyRelated Event
Title
Event type
ConferenceDate
09/11/2015 - 12/11/2015Location
Abstract
Model predictive control (MPC) technologies could deliver numerous benefits for power electronic converters and their applications. However, a varying degree of effort is required to integrate such technologies into industrial products. Moreover, the degree of complexity associated with the problem formulation and associated constraints, is linked to a number of unclear benefits for the industrial acceptance of such powerful yet inadequately appreciated control techniques by industry engineers. Translating the benefits of taking into account increased input control parameters for a more accurate and robust control against the classical single-input single-output Proportional-Integral-Derivative (PID) control is a serious challenge when one wishes to integrate such technologies in industrial products. The paper discusses MPC development steps across the spectrum of research, technology development and product development to contribute to the understanding of the challenges that need to be addressed to adopt such technologies into industrial products. The case of Medium-Voltage (MV) motor drives and High-Voltage Direct Current (HVDC) power transmission are considered in the paper. Both steady state and dynamic performance benefits are considered in the discussion and a number of potential paths to address the still limited commercial acceptance of such technologies are presented.
