River Water Quality Prediction Model Based on PCA-APSO-ELM Neural Network

The complex correlation and nonlinearity between various parameters of river water quality greatly affect the accuracy of water quality prediction. To improve the accuracy of dissolved oxygen prediction, a river water quality dissolved oxygen prediction model based on principal component analysis, improved particle swarm optimization, and extreme learning machine is proposed. This model extracts the principal components of water quality parameters through principal component analysis, reduces data dimensionality, and eliminates the complexity and correlation between water quality parameters; Introducing nonlinear dynamic adjustment of inertia weights, adaptive learning factors, and adaptive cross mutation strategies to improve the particle swarm algorithm, enhancing its optimization ability, avoiding falling into local optima, and finding the optimal solution faster; At the same time, an improved particle swarm optimization algorithm is used to iteratively optimize the input layer weights and hidden layer thresholds of the limit learning machine, avoiding the randomness of the initial input layer weights and hidden layer thresholds, improving the prediction accuracy of dissolved oxygen, and ultimately establishing a PCA-APSO-ELM neural network model for predicting dissolved oxygen in river water quality. This article takes water quality monitoring data from a certain section of the Haihe River basin as the research object. Compared with ELM, PCA-ELM, and PCA-PSO-ELM models, the PCA-APSO-ELM model has higher prediction accuracy, stronger generalization ability, and better prediction performance.