Building cooling and heating load accurate prediction method based on double attention mechanisms and novel particle swarm optimization

The invention relates to a building cooling and heating load accurate prediction method based on a double attention mechanism and a novel particle swarm algorithm, and belongs to the technical field of load prediction. The building cooling and heating load accurate prediction method comprises the steps that building historical cooling and heating load data and related external influence factor data are acquired, and the acquired data are preprocessed; establishing and training a BiLSTM (Bidirectional Long Short Term Memory) time sequence load prediction model according to the preprocessed data; introducing a double attention mechanism to carry out weight distribution on the input features and the hidden state of the neural network; performing hyper-parameter optimization on the load prediction model by using a novel particle swarm algorithm; establishing an error correction model by using a CEEMD decomposition technology and a BiLSTM neural network; and performing performance evaluation on the model according