With the rapid increase in the penetration rate of renewable energy

the randomness of power supply and load of the power system have superimposed on each other

increasing the difficulty of mid-long term power consumption balance and causing power shortage and renewable energy power abandonment to occur intertwined. The above issue poses a great challenge to the security and reliability of the power system. Therefore

this paper proposes a new mid-long-term production simulation method based on probabilistic sequences and approximate dynamic programming (ADP). First

this paper divides the entire year into multiple periods

either monthly or weekly. Before the start of each period

decisions regarding the power system operation mode need to be made. Considering source-load uncertainty

probabilistic sequence theory is used to calculate the system operating state for the next period

and a Markov decision process (MDP) is created for the power grid operation. Third

ADP is introduced into MDP to search for the optimal decision at different periods

and grid operation mode can be obtained. The new scheme avoids excessive reliance on heuristic principles and scenario selection in traditional methods

and it can significantly reduce the total cost while ensuring power supply safety and renewable energy consumption in extreme scenarios. Finally

the effectiveness and progressiveness of the proposed method are verified in the testing system.