Impact of the Stochastic Behaviour of Intermittent Renewable Resources on Power System Reliability

The widely acknowledged benefits offered by integration of Renewable Energy Resources (RERs) into conventional power systems have driven the development of these technologies and will continue to do so in future electricity networks, i.e. so called Smart Grids. However, there is significant uncertainty about the impact of these RERs on network reliability performance due to their inherent intermittence, unpredictability and spatial variability. Given the security and quality of supply (SQS) legislation, as well as the commercial implications of supply interruptions to stakeholders in the electricity sector, there is a need to quantify the probabilistic impact that different RER schemes will pose to network performance. Reliability indicators will quantify the risk of network outages, especially if insufficient preventive measures are adopted.

In this research, a test distribution system is modelled, scripted and fully controlled through Python with PSS/E software package. This system is designed based on typical distribution network design in the UK and incorporates the use of aggregation techniques to reduce not only the complexity but also the computational time associated with modelling of very detailed networks. By using both electrical and reliability equivalent models, an accurate distribution network model is built to facilitate the statistical analysis required to carry out the reliability analysis of a realistic, and therefore large, network. Accordingly, the presented research further develops the conventional Monte-Carlo Simulation technique by including the time-variation of network components’ failure rates and electricity demand profiles, as well as utilising a theoretical interruption model (Fig. 1) for assessing more accurately the moment in time when interruptions to electricity customers are likely to occur. Coupled with a comprehensive database of power components from UK DNOs, this analysis develops a more accurate reproduction of the stochastic nature of network performance. This facilitates a realistic base case upon which to design, deploy and assess different sets of RER schemes, by utilising energy storage systems to compensate for the intermittence of RERs as well as tailored demand-side response techniques (in part, based on the aforementioned theoretical interruption model).

The outcomes of this analysis demonstrate that intelligent deployment of RERs improves reliability by reducing, among others, the duration of supply interruptions (Fig. 2) and the energy not supplied to customers. Furthermore, the risk assessment of interruption times affecting domestic and non-domestic electricity consumers, subject to the regulator-imposed security of supply requirements, is quantified. Given that this is directly related to the associated cost of implementing preventive measures over corrective ones, this analysis provides insight into the commercial consequences resultant from RER integration. This is vitally important for prudent planning and operation of future electricity networks.