SOM-Based Robust Scheduling of Power Grid after Scenario Boundary Reduction

With the development of renewable energy sources, the integration of photovoltaic power generation in the power grid is increasing, posing challenges to the stability of the power grid. Ensuring the stability and safety of the power grid with uncertain energy sources and improving the economic efficiency of power grid scheduling are hot topics of current research. This study establishes an optimization model for multiple interconnected parks involving photovoltaic, wind power, and hydrogen storage systems, incorporating the impact of photovoltaic and wind power output, temperature, and pressure on hydrogen storage systems into uncertainty for robust linear optimization. Considering that the distribution of uncertainty has a certain probability bias, this paper uses the SOM neural network to cluster uncertainty scenarios and adjust the boundary information of robust optimization, taking into account the probability density of scenario distribution to a certain extent, reducing the impact of individual extreme rare scenarios on economic efficiency, thereby improving the overly conservative problem of robust optimization.