In order to eliminate the interference of PV output fluctuation on energy storage configuration
a method based on a PV output prediction model and an improved genetic algorithm (IGA) was proposed to determine the optimal energy storage capacity. Based on long short-term memory neural network (LSTM) to predict PV output
variational mode decomposition (VMD) to capture PV fluctuation characteristics
and improved sparrow algorithm (ISSA) to optimize LSTM neural network hyperparameters
a VMD-ISA-LSTM PV output prediction model was established. The optimal allocation of energy storage capacity is realized by an IGA that employs population partitioning and an improved mutation operator. The average absolute error
root mean square error and average absolute percentage error of PV output predicted by VMD-ISSA-LSTM can reach 12.48 kW
14.44 kW and 2.16%
respectively. After the improvement
the average convergence speed of genetic algorithm is increased by 51.33%
and the algorithm runtime is reduced by 31.16%. Considering the optimization of photovoltaic output
the installed cost of energy storage capacity is reduced by 6.63%
and the investment payback period is reduced by 0.7 years. The results of the example show that the method is helpful to solve the problems of large investment in energy storage construction
low operating income
long investment recovery cycle and other poor economic benefits caused by photovoltaic output fluctuations
and is conducive to promoting the sustainable and healthy development of photovoltaic energy storage industry and ensuring the smooth realization of the “dual carbon” goal.
ZHANG D L, MA Y Y, LIU J X, et al.Stochastic optimization method for energy storage system configuration considering self-regulation of the state of charge[J]. Sustainability, 2022, 14(1): 553.
ALQUTHAMI T, RAVINDRA H, FARUQUE M O, et al.Study of photovoltaic integration impact on system stability using custom model of PV arrays integrated with PSS/E[C]//North American Power Symposium 2010. Arlington, TX, USA, 2010: 1-8.