Author(s)

JinYi Luo

Affiliation(s)

School of Mathematical Science, Chengdu University of Technology, Chengdu 610059, Sichuan, China.

Corresponding Author

JinYi Luo

ABSTRACT

In recent years, the problem of water pollution has become increasingly serious, and traditional water quality monitoring methods are difficult to meet the demand for high precision. Constructing an efficient and reliable water quality prediction model is of great significance for governance decisions. To this end, this paper proposes a random forest (IPSO-RF) water quality prediction model optimized based on the improved particle swarm algorithm. Firstly, for the problems of traditional particle swarm algorithm (PSO), such as easy premature convergence and insufficient global search ability, an improved particle swarm algorithm (IPSO) with nonlinear iteration incorporating inertia weights is proposed, and its optimization performance is verified based on typical test functions. Secondly, the IPSO algorithm is combined with random forest (RF) to optimize the key hyperparameters of RF (e.g., the number of decision trees, the minimum number of samples for node splitting, etc.) in order to improve the prediction accuracy and generalization ability of the model. Simulation experiments were carried out based on the water quality monitoring data of a watershed, and the results showed that compared with RF and standard PSO-RF and other models, the IPSO-RF model showed lower MSE in the prediction of key indicators such as dissolved oxygen, phosphorus content of water body and ammonia nitrogen amount of water body, which verified its superiority in the prediction of water quality. This study not only provides new ideas for the application of intelligent optimization algorithms in the field of water environment, but also provides more accurate technical support for water quality monitoring and pollution prevention and control of environmental protection departments.

KEYWORDS

Random forest; IPSO-RF; Inertia weight; PSO

CITE THIS PAPER

JinYi Luo. Intelligent prediction of water quality parameters based on improved particle swarm optimization coupled with random forest model. Frontiers in Environmental Research. 2025, 3(2): 55-62. DOI: https://doi.org/10.61784/fer3030.

REFERENCES

[1] Rajaee T, Khani S, Ravansalar M. Artificial intelligence-based single and hybrid models for prediction of water quality in rivers: A review. Chemometrics and Intelligent Laboratory Systems, 2020, 200: 103978.

[2] Hmoud Al-Adhaileh M, Waselallah Alsaade F. Modelling and prediction of water quality by using artificial intelligence. Sustainability, 2021, 13(8): 4259.

[3] Chen Y, Song L, Liu Y, et al. A review of the artificial neural network models for water quality prediction. Applied Sciences, 2020, 10(17): 5776.

[4] Lu H, Ma X. Hybrid decision tree-based machine learning models for short-term water quality prediction. Chemosphere, 2020, 249: 126169.

[5] Jain M, Saihjpal V, Singh N, et al. An overview of variants and advancements of PSO algorithm. Applied Sciences, 2022, 12(17): 8392.

[6] Genuer R, Poggi J M, Genuer R, et al. Random forests. Springer International Publishing, 2020.

[7] Deepa B, Ramesh K. Epileptic seizure detection using deep learning through min max scaler normalization. Int. J. Health Sci, 2022, 6: 10981-10996.

[8] Choudhary S, Sugumaran S, Belazi A, et al. Linearly decreasing inertia weight PSO and improved weight factor-based clustering algorithm for wireless sensor networks. Journal of ambient intelligence and humanized computing, 2023: 1-19.

[9] Chen T, Chen X, Chen W, et al. Learning to optimize: A primer and a benchmark. Journal of Machine Learning Research, 2022, 23(189): 1-59.

[10] Chicco D, Warrens M J, Jurman G. The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation. Peerj computer science, 2021, 7: e623.