基于BPNN算法的浑河突发污染动态预警研究

Dynamic Early Warning Study on Hun River Sudden Pollution Based on BPNN Algorithm

频发、突发的水污染直接给环境带来严重的危害,引起相关部门与环境专家的重视,并展开对水污染预警响应技术的研究。根据国内突发性污染出现的问题,结合常规高频连续监测数据,基于BP神经网络相关技术对突发污染预警进行研究,与传统的预测技术相对比。该技术的预测精度更高,实用性更强,其原因是该技术能够去除单点异常带来的噪声干扰,对水质序列的突变更加敏感,为突发污染动态预警的相关研究提供理论参考。

For the frequent and sudden water pollution, it brings serious harm to the environment, attracts the attention of relevant departments and environmental experts, and conducts research on water pollution early warning response technology.According to the problems in domestic sudden pollution, combined with the conventional high-frequency continuous monitoring data, the BP neural network related technology is used to study the sudden pollution warning, which is compared with the traditional prediction technology. The prediction accuracy of this technology is higher and the practicality is stronger.The reason is that the technology can remove the noise interference caused by single point anomaly, and is more sensitive to the sudden change of water quality sequence, and provides a theoretical reference for the related research of sudden pollution dynamic early warning.