Prediction of SO2 Emission Concentration in Thermal Power Plants Based on SSA-LSTM Neural Network

In order to improve the prediction accuracy of SO
2
emission concentration, a SO
2
concentration prediction model based on SSA optimized LSTM neural network is proposed for SO
2
pollutant emission from coal-fired units in power plants. By using the combination of SSA and LSTM neural network, the optimal position of threshold update is quickly found in the local search to provide better parameters for the training of LSTM. The SSA-LSTM concentration prediction model is constructed using Python and simulated to compare with the LSTM prediction model, and the test results show that the SSA-LSTM neural network model is utilized to predict the concentration of SO
2
emission with good stability, and fast convergence, and the prediction accuracy is higher compared with the LSTM neural network model.