Economic Dispatch for Grid-Connected Wind Power With Battery-Supercapacitor Hybrid Energy Storage System

This study demonstrates an effective dispatching scheme of utility-scale wind power at one-hour increments for an entire day with a hybrid energy storage system consisting of a battery and a supercapacitor (SC). Accurate forecasting of wind power is crucial for generation scheduling and economic operation. Here, wind speed is predicted by one hour ahead of time using a multilayer perceptron Artificial Neural Network, which exhibits satisfactory performance with good convergence mapping between input and target output data. Furthermore, an adaptive neuro-fuzzy inference system is employed to devise a state of charge (SOC) controller to accurately estimate the grid reference power (P$_\mathrm{Grid,ref}$) for each one-hour dispatching period. This type of desired P$_\mathrm{Grid,ref}$ estimation is critical to ensure the energy storage system (ESS) completes each dispatching period with its starting SOC and has adequate capacity available for next-day operation. Also, the particle swarm optimization technique is implemented to optimize the life cycle cost of the ESS based on its depth of discharge usage, which is vital for minimizing the cost of a dispatchable wind power scheme. The actual wind speed data of four different days as a representative of each season recorded at Oak Ridge National Laboratory are utilized in the simulations to provide a realistic economic assessment for dispatching the wind power.