Refined Optimal Scheduling Model and Application for Thermal Power Units in High Penetration Renewable Energy Power System

Deep and start-stop peak regulation of thermal power units have become crucial measures to enhance the flexibility of modern power systems and facilitate renewable energy integration. Addressing the current limitations in optimization dispatch models that only consider binary operating states (online/offline) for thermal units, this study refines the scheduling modeling framework by comprehensively characterizing various operational states, peak regulation ranges, and start-stop conditions. A refined linear optimization dispatch model is developed, incorporating: compensation costs for deep peak regulation, segmented ramping rate constraints, start-stop operations under different operating modes, power output during start-stop transitions and reserved ramping capacity for ancillary services. This model can directly improve security constrained unit commitment algorithms in existing energy management systems. Case studies based on a provincial power grid with high renewable penetration demonstrate that the proposed day-ahead unit commitment strategy achieves integrated optimization of deep peak regulation capability, unit start-stop scheduling, power output planning, reserve capacity allocation and ramping capability management. The results indicate significant reductions in frequency stability risks caused by insufficient ramping capacity and improved renewable energy accommodation.