Thesis

A simulation-based optimisation method to evaluate dynamic compensators for the improvement of LCC-HVDC performance in high source impedance power systems

Creator
Rights statement
Awarding institution
  • University of Strathclyde
Date of award
  • 2024
Thesis identifier
  • T16970
Person Identifier (Local)
  • 201553477
Qualification Level
Qualification Name
Department, School or Faculty
Abstract
  • Line commutated converter (LCC) high voltage direct current (HVDC) link dynamic performance is negatively affected by low alternating current (AC) system short circuit ratio (SCR) as viewed from the LCC-HVDC link converter stations. This is particularly evident at LCC-HVDC link converter stations operating as inverters subjected to large transient disturbances. Firstly, this thesis proposes a simulation-based optimisation method to evaluate black-box optimisation solver methods built with mature strategies against alternative solver methods using surrogate model strategies recently proposed in the research literature. The method uses the problem of tuning LCC-HVDC link controllers considering large transient behaviour modelled via electromagnetic transient (EMT) simulations as the underlying motivating problem on which the solver methods are tested. The results from the applied method confirm the suitability of applying the tested surrogate-based solver methods on LCC-HVDC link controller tuning problems. The surrogate-based solver methods’ performances are improved between 45.137% and 72.14% relative to the worst performing solver method using mature strategies. Secondly, this thesis proposes a method to quantitatively evaluate dynamic compensators’ ability to improve the dynamic performance of LCC-HVDC links inverting into low SCR AC systems. The method uses EMT simulations as part of a simulation-based optimisation using one of the aforementioned surrogate-based optimisation solver methods to make fair comparisons between different compensator types and compensator ratings. Multiple inverter system short circuit fault locations and inverter system equivalent source impedances are considered in the method. Compensators are evaluated by performance values calculated via performance functions applied to measured time domain variable results from the simulations. The method is able to successfully quantify and differentiate compensator type and rating superiorities when applied to a set of static VAr compensator (SVC), static synchronous compensator (STATCOM), and synchronous condenser study cases. In particular, the method results show that any type of compensator of any rating typically improves LCC-HVDC link dynamic performance compared to a compensator-less LCC-HVDC link. The best found improvement is 9.2035% relative to the Base study case for the integral square error (ISE) of direct current (DC)-side measured power of the LCC-HVDC link. The method results also show that synchronous condensers are the most effective compensator, with improvements between 7.5269% and 9.2035% relative to the compensator-less LCC-HVDC link when considering ISE of DC-side measured power. Similarly, SVCs provide improvements between 5.4759%, and 5.7968%, and STATCOMs provide improvements between -0.21144% and 6.9608%. Smaller-rated SVCs and STATCOMs provide better improvements compared with larger-rated SVCs and STATCOMs, using the compensator-less LCC-HVDC link as a baseline.
Advisor / supervisor
  • Xu, Lie
Resource Type
DOI

Relazioni

Articoli