New reliability criteria for power system management
Power system reliability management allows the electricity grid to maintain performance at a desired level, while minimising the socio-economic costs of maintaining that performance level. In Europe, network reliability management has traditionally relied on the so-called ‘N-1’ criterion. This means that if one relevant element fails, the elements remaining in operation must be able to accommodate the new operating conditions without violating the networks’ operational security limits. Today, the increasing uncertainty of generation due to intermittent renewable energy sources, combined with the possibilities of demand-side management and energy storage, requires a new reliability criterion. The GARPUR project has responded to this challenge by designing, developing and assessing a new probabilistic reliability criterion for the current pan-European power system and its future development. Consequences of power system failures evaluated Researchers evaluated the relevance of the criteria and their practical use, while seeking to maximise social welfare. “The methodology is based on two novel elements,” explains project coordinator Oddbjørn Gjerde. “Firstly, it considers both the probability and the consequences of something going wrong in the power system. Secondly, it takes into account the socio-economic impact of that failure.” This new methodology allows TSOs to evaluate the probability and consequences of failures in their power system, which were expressed as the potential cost of power cuts to consumers. “With this information, TSO’s can make decisions and investments that provide the best balance between power supply security and cost,” explains Gjerde. The new criteria were devised with system development, asset management and system operation in mind, ensuring a consistent treatment of reliability across all time horizons. Mathematical and computational models were also used to predict the location, duration and level of interruptions to the power supply. The novel approach was designed to seamlessly fit with existing TSO processes. The TSOs tested the methodology for planning or operating their grids and the results were very promising. Getting the right balance The TSO partner in France compared the current N-1 rule with the new approach, and concluded that it would result in similar reliability at a lower cost. In Belgium the TSO assessed the methodology showing how clustering techniques can be applied to create operating states for a future year. The Icelandic TSO implemented probabilistic reliability assessment in the context of system operation, providing real-time risk visualisation and indicators valuable to the operators in the control room. In addition, the Norwegian TSO partner employed the methodology to conduct a system expansion planning study, which revealed that for a certain area it would be not socio-economically beneficial to choose the alternative with the highest reliability, since the increased investment costs outweigh the benefits. Major savings of about 25 % could be achieved in this case, mainly by lower investments cost. GARPUR will extend probabilistic reliability management from experts in the TSO organisations, who have the practical responsibility to ensure the security of electricity supply, to all stakeholders dealing with electric power systems reliability management. This also includes regulators and governments whose responsibility it is to ensure the system performs for the benefit of all parts of society. One of the most important benefits arising from GARPUR is that TSOs can now make better informed decisions. By applying the project’s methodology, the risk (defined as probability x consequence) of power cuts is known, allowing different alternatives to be compared, enabling optimal selection. “The aim is to find the optimal balance between the costs of providing reliable electricity supply and the socio-economic costs of power cuts,” concludes Gjerde.
Keywords
GARPUR, reliability, transmission system operators, power system, probability
