Air quality prediction based on improved quantum long short-term memory neural networks

Accurate prediction of urban air quality is of vital importance in preventing urban air pollution and improving the quality of life of urban residents. In order to achieve more accurate prediction of air quality, this study proposes a novel hybrid quantum neural network prediction model that combines an improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) and a quantum long short-term memory network (QLSTM) optimized by the particle swarm optimization (PSO) algorithm.This study focuses on using the QLSTM model to mine the time-real fluctuations and historical dependence of air quality data and applying the PSO algorithm to optimize the hyper-parameters of the quantum model to improve the prediction accuracy; then, ICEEMDAN is introduced to disassemble the original air quality data series into multiple pattern components containing different information, which effectively reduces the complexity of the data; the feasibility and validity of the proposed methodology are verified through comparison experiments with other prediction modeling methods. The results show that the proposed QLSTM prediction method incorporating ICEEMDAN and PSO optimization has the highest prediction accuracy in terms of prediction accuracy, which contributes a novel and quantum-specific technical approach to the field of air quality prediction.