Implementation of a Two‐Layer Architecture Frequency Regulation in Industrial Parks Using a Fuzzy Rule‐Based Neural Network

The large‐scale integration of renewable energy into the power system leads to volatility and intermittency, which may reduce the system's frequency control performance. However, model‐based methods struggle to address the uncertainty caused by renewable energy. Therefore, this paper considers the impact of renewable energy and designs a frequency coordination control scheme for industrial parks based on fuzzy rule‐based neural networks. Firstly, a frequency regulation model for industrial parks involving multiple demand‐side loads are established. Secondly, a primary frequency regulation–secondary frequency regulation two‐layer frequency regulation strategy utilizing a fuzzy rule‐based neural network is proposed to manage resource allocation of demand‐side loads participating in frequency regulation. This approach achieves faster, more reliable frequency regulation and increases the proportion of load resources involved in frequency control, and improves the economic efficiency of the industrial park. Finally, case studies are conducted in an industrial park with renewable energy sources such as wind and solar power integrated alongside electric vehicle loads, aluminium smelting loads, heating, ventilation and air conditioning loads, and electric heating boiler loads. The simulation results demonstrate that the proposed scheme improves system frequency stability and regulation speed by adjusting the power of controllable loads participating in frequency regulation.