Integrating component condition in long-term power system reliability analysis

Abstract

<p>The electric power system is a critical infrastructure in which power transformers play a key role in linking together generation and end-use of electricity. The consequences of transformer breakdown can be significant, and aging transformers have a higher probability of failure. For decisions in asset management and power system development, it will therefore be useful to capture how deteriorating component condition affects failure probabilities and the overall reliability of the power system. Since such decisions have planning horizons of multiple years, the analysis should also capture similar time horizons. To this end, this paper proposes an analytical approach to power system reliability analysis (PSRA) accounting for time dependencies in the technical condition of components. An analytical PSRA methodology integrating a transformer condition model is extended to analysis horizons of multiple years. This analytical methodology is compared with a Monte Carlo simulation (MCS) approach to PSRA by applying both to a realistic case study. The comparison validates the analytical approach by showing that the inaccuracies its approximations introduce are negligible, at least for the considered case. This means that the proposed methodology can be a computationally viable alternative to MCS methods, especially when it is too time consuming to assess the impact of different scenarios with sufficient statistical precision using MCS. However, drawbacks with the analytical approach for further extensions of the methodology are also discussed.</p>

Estimating Uncertainty in Reliability of Electricity Supply Analyses Considering Component Condition

“…where , and , denote equivalent failure rates and outage times for contingency j, and , , , denotes the power interrupted at delivery point k. For other details of the PSRA method we refer to (Toftaker and Sperstad 2022).…”

“…This section gives a brief recap of the component reliability model introduced in (Toftaker, Foros, and Sperstad 2023) and (Toftaker and Sperstad 2022). The main purpose of the reliability model is to include technical condition of individual power system components in the system reliability analysis.…”

“…To extend the component reliability model to a longer time horizon than one year we adopt the recursive scheme introduced in (Toftaker and Sperstad 2022). The expected number of wearout failures for component in year is given by the law of total expectation as…”

“…Integration of component condition in system reliability has been the topic of several research articles, e.g., (Li 2002) which handled end of life failures for transformers or (J. H. Jürgensen, L. Nordström, and P. Hilber 2019) which handled condition dependent repairable failures. Recent work (Toftaker, Foros, and Sperstad 2023) has shown how one may account for both condition dependent failures and preventive replacement for transformers in PSRA and how this can be carried out over an extended time horizon of several years (Toftaker and Sperstad 2022).…”

“…This paper builds on the work in (Toftaker, Foros, and Sperstad 2023) and (Toftaker and Sperstad 2022), which presented estimates for condition-dependent reliability indices but no estimate for their accuracy. A novel method to estimate the uncertainty in reliability indices is thus presented.…”