藻类生长预测模型训练的原始数据集（2022年）

通过对典型饮用水源地的藻类、水质、气候等数据进分析，建立藻细胞浓度与关键环境因子间的关系，确定关键参数值，然后利用logistic模型对未来2-4周的藻类的生长进行预测。同时，利用实际的藻类数据对模型进行连续的校正与优化学习，使得模型在长时间运行后，其预测精度逐渐增加高。通过对藻类的预测与实际对比结果发现：在为期180天的实际运行过程中，该藻类生长预测模型能较好的预测饮用水源地不同时空分布上的藻类情况，且其预测能力逐步增加。By analyzing data including algae, water quality, and climate from typical drinking water source areas, this study establishes the correlation between algal cell concentration and key environmental factors and determines the values of critical parameters. Subsequently, the logistic model is employed to predict algal growth over the next 2 to 4 weeks. Meanwhile, actual algal data is utilized for continuous calibration and optimization of the model, enabling its prediction accuracy to gradually improve with prolonged operation. Comparative analysis between predicted and actual algal outcomes reveals that during the 180-day actual operational period, the proposed algal growth prediction model can effectively forecast algal conditions across different temporal and spatial distributions in drinking water source areas, with its predictive performance improving progressively.

搜集汇总

背景与挑战

背景概述

该数据集是2021年9月至2022年9月期间在河南省信阳市浉河区南湾水库收集的原始数据，用于训练藻类生长预测模型。数据集包含生物量、叶绿素和水质、气候等环境因子，通过logistic模型预测未来2-4周的藻类生长，并在实际运行中进行校正优化，以提高预测精度。