Enhancing Power Grid Resilience to Earthquakes Using Defensive Algorithm

Abstract

Enhancing power resilience of electrical systems to natural hazards, such as earthquakes, windstorms, floods,etc. is crucial for avoiding disruptions in global infrastructures as well as for maintaining economic stability and public welfare. Earthquakes are one of the corruptive hazards that refer to High Impact Low Probability (HILP) events and can severely damage power stations, leading to massive outages and cascading failures. Thus, this research aims to raising the issue via introducing an innovative approach such as utilising a defensive islanding algorithm to enhance the seismic resilience of power networks. The approach encompasses different analytical tools, including but not limited to fragility curve & load flow analysis, spectral clustering, a Severity Risk Index employment, and Monte Carlo simulations to establish a comprehensive framework for risk assessment. Fragility curves are located at the epicenter when it comes to assesing the likelihood of component failure across various earthquake intensities and also it provides a comprehensive insight into system vulnerabilities. Peak Ground Acceleration (PGA) plays the role of the key variable for our analysis. Multiple earthquake scenarios are generated by Monte Carlo simulations to cover the stochastic characteristics of seismic hazards and their possible impacts on the electrical grid. A Severity Risk Index identifies the most susceptible branches and essential components of the network that may lead to power outages. For this introductory research of applying defensive islanding particularly to earthquakes, parameters mentioned above are more that sufficient for formulating a precise and well-defined strategy. Defensive islanding, which presumably divides the electrical system into autonomous sections, so called “islands” is a central component of this technique. It helps to mitigate interruptions during a seismic event or series of similar hazards. Additionally, spectral clustering is selected at the most recent & eficient techniques to improve the islanding process, making sure that each island is resilient and able to feed substantial loads. Overall, the strategy is helpful due to the fact that it enhances the operational capacity of the system not only during a hazardous event but also following that event. It divides the zones that are under higher risk and maintains the stability of unaffected areas. The methodology is defined through the test case study named IEEE 24-bus system which can also be reffered as “Reliability Test”. The MATPOWER library is mainly used for incorporating load flow analysis, evaluating the system's initial as well as subsequent characteristics. There are numerous variables analysed through MATPOWER: voltage levels, phase angles, power losses, etc. Simulation results highlight the fact that the strategy ultimately reduces the probability of cascade failures and improves the system's resistance to natural and sudden disturbances. The paper emphasizes the vital significance of resilience strategies in protecting fundamental loads during natural catastrophes, thereby providing necessary data to utility operators, researches and policymakers. The implemented strategy concentrates on a substantional improvement in the planning of seismic resilience for power systems. The research presents a pragmatic and dynamic methodology for mitigating earthquake hazards through the integration of probabilistic risk assessment and sophisticated network optimisation algorithms. This paper not only addresses power grid resilience but also offers effective ways for improving infrastructure stability and facilitating long-term recovery after seismic occurrences. Additionally, the study provides further room for improvement by incorporating real-time monitoring systems, IoT technology and machine learning models for early detection of natural hazardous events.