Lifecycle-Aware Optimization of Power Grid Projects for New-Type Power Systems: A Hybrid Approach Using NSGA-II and Hidden Markov Models

With the rapid evolution of new-type power systems, transmission and substation projects face increasingly complex challenges in multi-objective trade-offs, lifecycle transitions, and operational uncertainties. To achieve intelligent project selection and dynamic lifecycle management, this paper proposes an integrated decision-making framework that combines the Nondominated Sorting Genetic Algorithm II (NSGA-II) with a Hidden Markov Model (HMM). The proposed method features a threelayer structure composed of project data management, NSGA-IIbased multi-objective optimization, and HMM-based lifecycle state tracking. Specifically, the optimization module identifies a Pareto-optimal project set by minimizing historical technical failure rates while maximizing supply assurance. The state tracking module infers the stage-wise probability distributions-Reserve, Planning, Construction, and Operation-based on timeseries indicators such as approval progress and equipment arrival rate, and it triggers early warnings when abnormal transitions are detected. To validate the model, a case study is conducted using 200 real transmission projects from a northern China power grid company. Simulation results demonstrate that the proposed approach effectively selects projects with balanced technical reliability and supply capabilities, achieves 91.6% accuracy in lifecycle state recognition, and enables early risk warnings more than one week in advance. These findings confirm the practical applicability and generalization value of the proposed method for intelligent management of infrastructure projects under new-type power system conditions.