Research Progress and Challenges in Phasor-Based Time-Domain Simulation Solution Algorithms for New-Type Power Systems
The large-scale integration of renewable energy sources has imposed strong nonlinearities and multi-timescale coupling on modern power systems, rendering conventional time-domain simulation solution algorithms inadequate for offline transient analysis of large-scale new-type power systems with high proportions of power electronic devices. This review systematically addresses the performance optimization problem of time-domain simulation algorithms. First, new requirements are identified across three solution stages: numerical integration, nonlinear algebraic equation solution, and linear algebraic equation solution. Second, existing research progress is consolidated under two core optimization paradigms—algorithm modification and algorithm adaptation—with their respective advantages and challenges analyzed. Algorithm modification pursues mathematical format innovation to expand absolute performance boundaries, while algorithm adaptation achieves optimal resource allocation within given boundaries through real-time scheduling and combination. The review concludes that the core challenge lies in coordinating multi-dimensional performance conflicts and responding to time-varying simulation demands. Persistent deficiencies include: in algorithm modification, multi-dimensional performance conflicts hinder reform efforts, diminish performance gains, and fixed algorithmic structures cannot respond to time-varying requirements; in algorithm adaptation, adjustment objectives and state-sensing dimensions remain singular, with parameters and switching logic heavily reliant on manual experience, severely constraining robustness. Future directions advocate integrating both paradigms to expand performance adjustment ranges and enhance responsiveness to time-varying demands, while exploring artificial intelligence fusion with traditional numerical methods to replace empirical parameter design and achieve precise algorithm regulation.