Dataset on Complex Power Systems: Design for Resilient Transmission Networks Using a Generative Model

Critical infrastructure systems are prone to be a target of disruptions from either natural disasters or deliberate attacks that can cause a great magnitude of negative consequences. To withstand and reduce the impacts of disruptive events, it is beneficial to consider them in the early stage of design to create a configuration that can maintain its performance as much as possible. However, due to the complexity of modern systems, it is difficult to optimize the large-scale system design with complicated constraints directly. To overcome this issue, it is advantageous to tackle the problem with a state-of-the-art data-driven approach. However, a subsequent problem arises due to the limited amount of data for power systems available to the public. Hence, synthesized power systems that resemble existing networks are collected and processed, which can be utilized for implementing machine-learning models. The power system networks are represented as graphs with features holding information on the nodes and edges that relate to electrical components. Using the generated dataset, this research implements a generative approach based on Wasserstein generative adversarial networks (GAN) to create intelligent designs. Afterward, performance measures are employed to induce the generation of a population with more appealing characteristics. To demonstrate the practicality of the dataset on data-driven design, it was applied to train the generative model for a 57-bus system and the trained generator was employed to an optimization framework. The results have shown that the model could generate feasible designs of higher resilience and the framework was capable of finding an optimal network design with improved performance.