Wide-Area Monitoring based Smart Frequency Control in Future Low-Variable Inertia Systems with CCGT/Wind/PV/BES/Load

UoM administered thesis: Phd

  • Authors:
  • Rasoul Azizipanah-Abarghooee


Renewable energy sources have been extensively developed in many countries like United Kingdom and academia like enhanced frequency control capability (EFCC) project in the last decades due to environmental and energy security risks, however, the system inertia is decreased and frequency response imposes a new challenge for power system operators to maintain power quality. Additionally, restricting RoCoF within satisfactory restrictions is so critical to prevent triggering protection relays that can result in cascaded problems and violate system security. The inertial response (IR) and droop based primary frequency response (PFR) using available spinning reserve power are inherently deployed in conventional power plants (CPPs) to restore sudden frequency deviations. However, power converter interfaces decouple the network frequency from the rotational part of wind energy conversion systems (WECS). In addition, the photovoltaic (PV) units have static dc generators. Thus, these generating units, by themselves, neither obtain the IR nor participate in PFR mechanism. A faster and smarter frequency response is required in order to compensate this phenomenon. The objective of the research presented in this thesis is to derive the local and wide-area smart frequency control based wide-area monitoring system (WAMS) supported by PMU-based local and wide-area monitoring systems, which is able of releasing fast, smart and robust primary frequency response from fast service providers like combined cycle gas turbine (CCGT), WECS, PV, battery energy storage (BES) and smart induction motors (SIMs). The research proposed in this thesis includes the creation of a novel robust process for the on-line estimation of loss of generation considering load damping impact, PMU latency and implausible spike in frequency data. Furthermore, the CCGT dynamic behaviour is accurately modelled and their unique fast frequency response following the frequency excursions is investigated. This research study also proposes a new wind unit control architecture based on inertial emulator and pitch controller to enhance its frequency response for its high integration in large-scale power systems. The novelty lies in the technique to defining the reference injection power from a control structure based on the novel maximum power point tracking (MPPT) surface deploying rotor speed and pitch angle which doesn’t rely on wind speed information. Additionally, the accurate drive train aerodynamic equations of wind turbine are also implemented beside the double-mass mechanical dynamics to investigate the oscillation frequency of electromechanical eigenvalues. Furthermore, a PV and vanadium redox BES enhanced with the control framework to provide primary frequency control, suitable for large scale frequency studies are provided. In case of PV, a primary controller is introduced which permits droop-control operation at deloaded MPPT based on available reserve power and emulated inertial power capturing from dc link capacitor by adjusting the dc link reference voltage. A frequency controller deploying a traditional droop based controller as well as inertial emulator regulates the vanadium redox battery’s active power transfer following the disturbance. It is notable that the inertial power is taken from dc link capacitor by adjusting the dc link reference voltage. In addition, a new speed controller is derived enabling the smart induction motor variable frequency drive systems to alter their consumption power in proportion to the grid frequency changes. With this aim, the reserve rotor speed of primary frequency controller is quantified accordingly and multiplied in a droop gain in order to create fast and controllable reserve power through power electronic interface. A rate limiter is also deployed to pose the rate of change of reserve rotor speed. The motor active power and load inertia are utilized to determine this rate for the first time. The motor inertial power is also suitably emulated. In order to evaluate the effectiveness of the proposed techniques and modelling, a large-scale 36-Zone Great Britain (GB) transmission system network reduction from National Grid (NG) and further data and developed modelling of all service providers is presented to support research into future power networks. The model is intended to be purely indicative of the topology, impedance characteristics driving inter-area and other phenomena across a network of such scale within which a variety of control approaches and dynamic performance characteristics of connected sources may be explored. Flexible architecture of test network permits a broad range of sensitivity studies and physical phenomena critical for design of future power networks.


Original languageEnglish
Awarding Institution
Award date1 Aug 2019