A novel groundwater burial depth prediction model-based on the combined VMD-WSD-ELMAN model.

Journal: Environmental science and pollution research international
Published Date:

Abstract

The improvement of groundwater burial depth prediction accuracy is an important guiding significance for the development and management of groundwater resources. Groundwater burial depth sequence has the characteristics of uncertainty and nonlinearity. Variational mode decomposition (VMD) has a powerful advantage in dealing with nonlinearity. Wavelet signal denoising (WSD) can reduce the high-frequency component noise so that its abrupt change point is reduced. Meanwhile, ELMAN neural network has the advantages of stability, adaptability to time-lapse, and dynamic memory. Based on their advantages, the combined VMD-WSD-ELMAN model is developed and applied to groundwater prediction in the People's Victory Canal Irrigation Area. To verify the reliability of the model, the prediction results were compared with the single ELMAN network and EMD-ELMAN model, and the results showed that the combined VMD-WSD-ELMAN model has higher accuracy and 100% qualification rate, and the prediction results are better than the single ELMAN model and EMD-ELMAN model. The model reveals the future spatial distribution of groundwater and its dynamic changes with time and provides a basis for future dynamic artificial numerical simulation.

Authors

  • Xianqi Zhang
    Water Conservancy College, North China University of Water Resources and Electric Power, Zhengzhou, 450046, China.
  • Dong Zhao
    Collaborative Innovation Center of Judicial Civilization, Key Laboratory of Evidence Science, Ministry of Education, China University of Political Science and Law, Beijing 100088, China.
  • Bingsen Duan
    Water Conservancy College, North China University of Water Resources and Electric Power, Zhengzhou, 450046, China.
  • Wenbao Qiao
    Water Conservancy College, North China University of Water Resources and Electric Power, Zhengzhou, 450046, China.