To enhance the precision of wind-solar power prediction, we propose an enhanced gray wolf algorithm (IGWO) that incorporates an improved adaptive factor and elite backward learning strategy. The aim is to optimize the long short-term memory network (LSTM) prediction model. By utilizing IGWO to optimize the parameters of the LSTM fully connected layer, we establish a combined IGWO-LSTM model for predicting wind-solar power generation. This model improves the traditional LSTM model's convergence speed and solving efficiency, while effectively avoiding local optimal solutions. Through simulations based on real-time data from a PV power station in Tianjin Eco-city, we demonstrate that the IGWO-LSTM outperforms the LSTM in accurately predicting PV power. Experimental analysis further confirms the effectiveness of our model. These examples demonstrate that our proposed method can effectively and accurately predict wind power generation system output, thereby enhancing power grid security and stability.
IGWO-LSTM based power prediction for wind power generation
11.10.2023
2756609 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Navigation reminding method for short-term traffic flow prediction based on IGWO-LSTM
Europäisches Patentamt | 2021
|Fault Type Identification in Transmission Lines based on VMD-FE and IGWO-KELM
British Library Conference Proceedings | 2020
|