Parallel Simulation Environment for Planning the Development of Electric Power Systems under Terrorist and Military Threats

The paper is devoted to the problem of protection and ensuring the reliability of electric power systems (EPS) on a national scale in the face of targeted threats. To address this problem, it is necessary to ensure structural variability of the energy system, which involves the use of different types of generating capacity in each regional energy subsystem, including renewable and advanced nuclear technologies, as well as energy storage systems. Their models are mathematically described by high-dimensional mixed integer linear programming (MILP) problems that require parallelization and HPC computing resources to reduce computation time to acceptable values. The goal of the research is to propose a general concept and develop a structure for a parallel multilayer environment for solving simulation tasks of structurally variable EPS under terrorist and military threats, suitable for addressing the challenges of restoration and optimal development of Ukraine's energy infrastructure during martial law and post-war reconstruction. The key components of the proposed framework were implemented and tested using the example of the problem of optimizing the structure of generation capacities in a local EPS as a MILP task. Free linear optimization solvers SCIP and HiGHS were tried. To parallelize the computational task, population-based approaches such as the Evolutionary Centers Algorithm (ECA) and Ant Colony Optimization (ACO) were applied. Some preliminary experimental results were obtained and analyzed. Directions for further development of the environment were substantiated.