Abstract:
Electrical power systems are undergoing innovative transformation of their existing practices to realize the smart grid (SG) vision. SGs are digitally empowered electrical grids that have the capability of controlling a large number of geographically spread generation, storage and loads across transmission and distribution networks. However, currently in operation and planning of networks, the interaction between increasing distributed generation (DG) and demand response (DR) present in the system is not explicitly factored. The development and installation of smart technologies such as intelligent electronic devices (IEDs) are increasingly happening in recent years. To protect and control this highly multi-layered system the SG needs to develop and establish robust protection and control schemes. Load shedding (LS) is one of the last actions that are automatically initiated, based on its designed operation, to prevent wide area system collapse when all other possibilities have been exhausted. Most of the existing LS schemes are well established at a transmission/sub-transmission level and executed by controlling the feeder load rather than individual loads downstream. In recent years traditional networks are shifting towards a smart grid framework where devices are equipped with emerging technologies and comply with standard based sub–station communication framework, such as International Electrotechnical Commission (IEC) 61850 Generic Object Oriented Substation Event (GOOSE) messaging. Operational control elements are also shifting from distribution to the individual interconnection level. These technological developments can potentially lead to the creation of a dynamic market for Automatic LS in the foreseeable future. Several catastrophic global blackouts have occurred, including New Zealand (NZ), in the last few years. Inefficient design of existing LS schemes is one of the critical reasons at times for resulting in larger blackout footprint. In the context of SG, emergency controls (including the LS) that are used to prevent blackouts need to be revisited. It is difficult to prevent system blackouts entirely. However, protection and control procedures can be improved by using the emerging technologies to help reduce the geographical span of blackouts. In the context of emerging technologies and sub-station communication standard based framework, this research aims to systematically propose developments of improved and newer LS strategies. In this thesis the major technical issues are addressed to improve LS mechanisms and the aim is to facilitate SG adoption in realize these techniques in practice. Existing LS schemes are first investigated to address the merits of each technique. Subsequently, a number of modifications for enhancing the existing LS techniques of each method have been outlined. This thesis introduces some new techniques which help to improve overall protection and control mechanisms. The proposed systems incorporate some important power system stability elements, especially voltage and frequency, by providing a real-time and optimal load control and shedding strategy, during a situation where the power system would otherwise have become unstable. The development of algorithms and procedures based on real-time distributed intelligence across the power system network, rather than the traditional centralised scheme, will be one major contribution of this thesis. These algorithms are intended to provide a flexible and decentralised method of mitigation. A secure implementation pathway also falls under the scope of this work which is approached using emerging substation standards like IEC 61850 and other open standards and implementation pathways. Thus, this thesis demonstrates that through SG technologies there are opportunities for improving upon traditional centralised LS schemes towards a more decentralised, smarter and graceful outage implementation. Overall, the developed algorithms, models and methodologies are tested for both transmission and distribution networks. To demonstrate LS technique in an SG framework, an experimental setup at the University of Auckland power systems laboratory, is also reported. To validate the proposed LS method, real New Zealand network data has been used for all the illustrative case studies. Keywords: Automatic Under Frequency Load Shedding; Automatic Under Voltage Load Shedding, Blackout, Distribution Network, Distributed Generation, Demand Side Management, Demand Response, Distribution System, Distribution Network, Energy Shedding, Emerging Technology, Intelligent Electronic Device, GOOSE Messaging, IEC 61850, Load Shedding, Load Reduction, Load Model, Load Characteristics Programmable Logic Controller, Power Generation, Power System Protection Relaying, Power System Faults, Power System Protection, Power System Stability, Rate of Change of Frequency; Smart Grid, Transmission System, Spinning Reserve, Under Frequency, Under Voltage.