A Hybrid Air Quality Prediction Model Based on Signal Decomposition and Multi-Objective Optimization Algorithm

Accurate and stable prediction of the Air Qual-ity Index (AQI) is a hot topic among many researchers. In response to the nonlinear oscillation of AQI data itself and the tendency for many studies to focus only on accuracy while neglecting stability, a hybrid prediction model based on intel-ligent optimization algorithms has been developed. The model utilizes Sequential Variational Mode Decomposition (SVMD) to reduce data complexity and optimizes Elman Neural Net-work (ENN) using a newly proposed multi-objective algorithm called Merged Differential Evolution and Whale Optimization Algorithm (MODEWOA). The algorithm employs two fitness functions that consider both accuracy and stability. Results show that the proposed MODEWOA algorithm outperforms MOPSO and NSGA-II, while the hybrid model surpasses other comparative models in terms of prediction accuracy and stability.