Decentralized Dynamic System for Optimal Power Dispatch in Wind Farms Based on Node-Dependence Nature

Abstract Meeting the power demand from transmission system operator (TSO) is an important objective for power dispatch, which introduces a power supply-demand constraint coupling all the wind turbines (WTs) among the wind farm (WF) into the optimization problem. Traditional methods, such as centralized or distributed optimal methods, can effectively handle global constraints from the TSO. However, these methods are unsuitable for real-time applications because of the high computational and communication requirements. In recent years, decentralized optimal control has emerged as a promising solution for WF control, but solving the coupling constraints is still challenging because of the lack of information sharing between local controllers. To this end, this study designed a completely decentralized dynamic system for WFs to optimize power dispatch while satisfying the TSO's power dispatch requirements. A voltage optimization problem with supply-demand constraints was decoupled based on node-dependence nature, an inherent characteristic of WFs that can be expressed using power sensitivities. The local optimization problem was solved iteratively using the gradient projection method, and the system converged linearly to the equilibrium point. Case studies in MATLAB/Simulink demonstrate that the proposed method can achieve near-global optimal performance only using local measurements.