Power system dynamics with increasing distributed generation penetrations

Abstract

Future power systems are likely to have a significant amount of renewable power generation due to favourable government policies, environmental issues and economic merits. Power systems are rapidly transitioning toward having an increasing proportion of generation from non-traditional inverter-based resources such as wind and solar power. An inevitable consequence of the power system transition towards nearly 100% inverter-based resources is the loss of synchronous generators with their associated inertia, frequency, and voltage control mechanisms. However, to ensure the future power system needs can be met, the source of system services to meet these needs will change from synchronous plants to other sources such as demand response and battery energy storage technologies. Demand response based on internet data centres is projected to become an increasingly important asset to contribute to ancillary service markets. In this dimension, Internet service companies are expected to combine the capabilities of a variety of data centre onsite resources to participate as a single provider similar to a virtual power plant.

Under this context, this research investigates three frameworks for data centres to deliver fast frequency response services, namely uninterruptable power supply, cooling units, and the ability to off-grid the entire data centre. These onsite flexible resources are modelled in DIgSILENT PowerFactory using dynamic and static providers, respectively. The performance of the proposed operational frameworks is validated inside the high fidelity 39 Bus system calibrated to an actual frequency event that occurred in the Irish power system. The sensitivity analysis demonstrates that both static and dynamic can significantly improve system frequency metrics and can arrest frequency nadir in the early stage of a disturbance. However, compared to the dynamic response, a substantial improvement is found in the system frequency when the number of static step responses decreases to withdraw a large amount of energy within the timeframe of inertial response.

Then, the data centre frameworks are further developed by incorporating delay-tolerant workloads and backup power supply units to provide a fast frequency response service. This is achieved by employing model predictive controllers that initiate reference signals to each data centre resource while respecting device operating conditions and constraints. Simulation results demonstrate the potential of different data centre configurations to quickly stabilise grid frequency under different wind penetration levels, during signal delays and severe cascade failures. The analysis shows that the proposed framework is critical to the adoption of renewable energy and reduces the requirement for an expensive spinning reserve used in a typical power system. However, it is shown that in low inertia power systems, the late response of the service not only deteriorates the system frequency metrics but can also result in complete system instability. Thus, an adaptive delay compensator is proposed to alleviate the impact of phase lag issues due to the time variant signals.

Finally, the analysis is extended to investigate the effect of various fast frequency response service locations and technology types on the system stability. Battery energy storage with different converter technologies is expected to address the challenges of displacing synchronous generators. This is because of their fast ramping capabilities and the ability to replicate functionalities that so far have been provided by conventional generators. To explore this capability, the potential benefits of battery energy storage equipped with grid following and grid forming converters are thoroughly investigated. Performance comparisons that account for the interactions between synchronous generators and converter technologies are also studied via dynamic simulations for the projected 90% non-synchronous inverter based resources in Ireland in 2030. The empirical findings demonstrate that inertia represents only one aspect of the power system’s needs and besides reduction of total system inertia maintaining network stability requires adequate system strength and grid impedance.

Note Appendices A.3 & C.5 have been redacted for educational development purposes 

Date of Award	Jul 2022
Original language	English
Awarding Institution	Queen's University Belfast
Sponsors	Collaborative REsearch of Decentralization, ElectrificatioN, Communications and Economics (CREDENCE) , Northern Ireland Department for the Economy, Science Foundation Ireland & National Science Foundation
Supervisor	Aoife Foley (Supervisor) & David Laverty (Supervisor)