A multi-timescale coordinated scheduling algorithm for zero-carbon power systems under high percentage of new energy grid integration

The power generation output of new energy is affected by weather conditions, seasonal changes and other factors, which exhibits high volatile and unpredictable, which makes the coordination and dispatching of zero-carbon power systems under the high proportion of new energy grid connected significantly more difficult. The multi-timescale coordinated scheduling approach divides the scheduling process into multiple time scales, with each time scale having distinct scheduling objectives and strategies. This hierarchical scheduling strategy effectively handles the uncertainty of new energy power generation output. Therefore, a multi-timescale coordinated scheduling algorithm for zero-carbon power systems under high percentage of new energy grid integration is proposed. Based on the load demand, power balance, and energy storage characteristics of the zero-carbon power system, the output of the generator units of the zero-carbon power systems under the high proportion of new energy grid connection is predicted. In light of the forecasted output of generator units, the model aims to minimize daily operating costs while adhering to constraints imposed by both the power and thermal systems, a day-ahead scheduling model for the zero-carbon power system is formulated and solved. Based on the day-ahead dispatching, the MPC algorithm is employed to develop an intra-day rolling dispatching model. This model uses the predicted and measured values of wind power and photovoltaic active power output of zero-carbon power systems after day ahead dispatching. This model leverages predicted and measured values of wind and photovoltaic active power outputs, post day-ahead dispatching to generate control sequences. Control variables are introduced to implement feedback correction for the model, so as to achieve multi-timescale coordinated dispatching of zero-carbon power systems. The experimental results show that the proposed algorithm outperforms the comparative methods in multiple key performance indicators: in terms of prediction accuracy, its power generation unit output prediction results are closer to the actual values; In terms of new energy consumption, the maximum photovoltaic consumption of the system is 8420 kW·h, which is significantly higher than the comparative method; In terms of system stability, the power supply–demand imbalance rate has always been below 1.71%, effectively improving the economy, flexibility, and operational reliability of zero carbon power systems under high proportion of new energy access, providing a feasible technical path for scheduling optimization of new power systems.