Application of genetic programming in electrokinetic treatment of soil and COPR: Focus on Cr(VI) fractionation

Created by W.Langdon from gp-bibliography.bib Revision:1.8414

@Article{Yu:2025:jece,

• author = "Qiu Yu and Yi Zheng and Zhao Jiang and Beichen Yu",
• title = "Application of genetic programming in electrokinetic treatment of soil and {COPR:} Focus on Cr({VI)} fractionation",
• journal = "Journal of Environmental Chemical Engineering",
• year = "2025",
• volume = "13",
• number = "2",
• pages = "115420",
• keywords = "genetic algorithms, genetic programming, Electrokinetic remediation, Cr(VI) removal, Artificial intelligence algorithm, Deep learning, ANN, Self-adaptive optimization",
• ISSN = "2213-3437",
• URL = "https://www.sciencedirect.com/science/article/pii/S2213343725001150",
• DOI = "doi:10.1016/j.jece.2025.115420",
• abstract = "Eliminating highly hazardous hexavalent chromium (Cr(VI)) from solid waste is momentous for alleviating Cr(VI) toxicity to the environment. Cost-effective electrokinetic remediation technology was used for the Cr(VI) removal from contaminated soil and chromite ore processing residue (COPR). While the technology's principle was straightforward, the associated physicochemical processes and pollutant characteristics were highly intricate, and it is hard to build the quantitative relationship between the treatment effect and various influencing factors. Thus, this study innovatively applied artificial intelligence algorithm to solid waste disposal, established a deep learning prediction model based on genetic programming (GP) for Cr(VI) evolution in water-soluble, weakly adsorbed, and insoluble form, and specifically analysed how the algorithm function enabled the model's self-adaptive optimisation. Results displayed that the generated prediction models were differentially distributed depending on the pollution medium and Cr(VI) fractionation. Current and anolyte pH were dominant in Cr(VI) elimination, and catholyte pH played a greater role in COPR than in soil. The predicted values was close to the experimental values, the prediction accuracies were all higher than 95percent, and could reveal the evolution patterns of Cr(VI) in different forms. For weakly adsorbed and insoluble Cr(VI) prediction in COPR, the error level was within 4percent and the data frequency within the error level of 4percent reached 90percent for water-soluble Cr(VI), suggesting that GP accurately predicted the majority of the Cr(VI) concentration data, which was an important guideline for optimising electrokinetic remediation schemes",

}

Genetic Programming entries for Qiu Yu Yi Zheng Zhao Jiang Beichen Yu

Citations