A Fully Distributed Optimization Method for Reserve Capacity of Multi-Area Interconnected Power Systems Considering Reserve Sharing

Large-scale grid-connected generation of renewable energy makes power grid reserve optimization more important. In this paper the reserve optimization for multi-area interconnected power systems (MAIPSs) with wind power is studied. Firstly, the uncertainty set of wind power output with budget constraints is constructed to represent the spatial correlation of wind power output. On this basis, the reserve demand of wind power is calculated by solving some simple linear programming problems. Then the total reserve demand for wind power and loads in all areas is considered synthetically, and an optimization model for reserve capacity of MAIPSs is established, which takes maximum wind power accommodation of the whole system as the optimization objective and considers reserve sharing and various operational constraints of the grid. Subsequently, combined with the characteristics of global optimization variables in the system, the existing distributed (sub)-gradient optimization algorithm based on consensus is improved to achieve a fully distributed solution of the model faster. The new global variable consensus calculation is proposed and additional global variable (sub)-gradient consensus operation is added, which ensures the convergence of the improved algorithm under fixed step size. Finally, the IEEE 24-bus system is divided into 3-area interconnected power systems to perform the numerical tests. The results show that the improved distribution optimization algorithm can obtain the global optimal solution more quickly, and the proposed reserve optimization model can improve the level of wind power accommodation.