The present invention provides a wind-storage
joint frequency regulation control method based on
model predictive control and
deep learning, comprising: constructing an
energy storage equivalent model suitable for wind-storage
joint frequency regulation application scenarios; constructing a
discrete system prediction model for wind-storage joint
system participating in
frequency regulation, predicting the
system state within the k+1 to k+Np steps, setting corresponding constraints and objective functions, offline traversing different
system load fluctuations, system frequency deviations, and
energy storage system SOC working conditions, respectively obtaining the optimal effective inequality constraint set, and
numbering the system working
modes according to the different inequality constraints satisfied; taking the system load fluctuation, system
frequency deviation, and
energy storage system SOC as inputs and the system working
mode number as output, training the DNN so that it can obtain the current optimal effective inequality constraint set according to the current system load fluctuation, system
frequency deviation, energy storage system SOC state and the predicted system load fluctuation, system
frequency deviation, and energy storage system SOC state in the predicted
time domain, and mapping the optimal power
instruction sequence; selecting the first one in the optimal power
instruction sequence as the wind-storage system power instruction and issuing it to the wind-storage system. The present invention can achieve the minimum system
frequency regulation cost while optimizing the
frequency regulation performance and system frequency stability, thereby improving the economic efficiency of wind storage system operation.