From baseload to flexibility: A cost-driven evolutionary model for thermal power industry at various stages of the China's “dual carbon” policy

Against the backdrop of China's “Dual Carbon”, the role of thermal power in high-renewable power systems is shifting from baseload to flexibility regulation, and its economic viability is crucial for the energy transition. This paper aims to reduce the operating costs of thermal power in various periods. First, a thermodynamic-data fusion approach quantified the impact of steam extraction on feasible operating ranges. Then, a multi-objective optimization model (based on genetic algorithm) integrating heat-power coupling characteristics is developed to dynamically allocate electrical and thermal loads among clustered units, which incorporates heat-consumption costs, ramp-wear costs, and start-stop costs. Finally, the cost-driven evolutionary model adaptable to 0–70 % renewable energy penetration rates is constructed, which can effectively schedule unit retirements. Results demonstrate that the optimized dispatch strategy reduces generation costs by up to 7.12 % compared to equal load distribution, with wind power integration showing greater cost advantages than solar power. Winter with highest heating demand, yields the lowest electricity cost prices (24.774 USD/MWh). The evolutionary model reduces generation costs by 3.06 %–4.20 % in high-wind and 1.24 %–2.06 % in high-solar penetration scenarios by retiring thermal units. The study provides actionable insights for policymakers and operators, and inspiration for virtual power plant scheduling.