Scholarlink

The cascade of water reservoirs is a complex and non-linear system with many unpredictable natural and human-dependent factors that define water flow characteristics and influence water quality. This work provides an analysis of radioactive pollution data collected in 1999 for all six reservoirs and major inflowing rivers. The regression techniques were applied to forecast water flow and future water contamination. The initial stage of research is to determine the optimal set of input parameters that balances the addition of new parameters with noise available in the dataset. The next stage is to find the best machine learning algorithm, and for this specific task, we selected an extra trees regressor. The final stage is to forecast the propagation of radioactive pollution through a cascade when only initial pollution impulse data are provided.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Forecasting the Dnipro Water Reservoir Cascade Pollution Using Machine Learning Methods

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Anatoliy Doroshenko,
  • Vladimir Sizonenko,
  • Dmitry Zhora,
  • Olena Yatsenko

摘要

The cascade of water reservoirs is a complex and non-linear system with many unpredictable natural and human-dependent factors that define water flow characteristics and influence water quality. This work provides an analysis of radioactive pollution data collected in 1999 for all six reservoirs and major inflowing rivers. The regression techniques were applied to forecast water flow and future water contamination. The initial stage of research is to determine the optimal set of input parameters that balances the addition of new parameters with noise available in the dataset. The next stage is to find the best machine learning algorithm, and for this specific task, we selected an extra trees regressor. The final stage is to forecast the propagation of radioactive pollution through a cascade when only initial pollution impulse data are provided.