利用信号处理预测藻类繁殖:从集体学习的新视角.
Caicai Xu1, Yuzhou Huang2, Ruoxue Xin3
1Institute of Zhejiang University-Quzhou, 99 Zheda Road, Quzhou 324000, China; Key Laboratory of Biomass Chemical Engineering of Ministry of Education, College of Chemical and Biological Engineering, Zhejiang University, Hangzhou 310027, China; Shandong Key Laboratory of Marine Ecological Environment and Disaster Prevention and Mitigation, Qingdao 266061, China.
Water research
|May 23, 2025
概括
通过将信号处理与机器学习相结合,可以提高精确的藻类繁殖预测. CEEMDAN-Hybrid-Ensemble (CHES) 模型提高了预测准确度和早期预警系统的稳定性.
科学领域:
- 环境科学 环境科学
- 数据科学数据科学数据科学
- 应用数学 应用数学 应用数学
背景情况:
- 准确预测藻类繁殖对于有效的管理和减缓策略至关重要.
- 独立模型与藻类开花的复杂时间频率动态作斗争.
- 现有的方法往往缺乏现实世界环境监测所需的稳定性.
研究的目的:
- 开发一个先进的集体框架,以改善藻类繁殖预测.
- 将信号处理技术与机器学习相结合,以提高预测准确度.
- 创建一个强大的模型来预测藻类动态在各种时间和空间尺度.
主要方法:
- 利用完整的集体实证模式分解与适应性噪声 (CEEMDAN) 来分解非静止藻类动态.
- 采用一组四个不同的机器学习模型从分解的组件中学习.
- 开发了CEEMDAN-Hybrid-Ensemble (CHES) 模型,该模型结合了信号处理和机器学习.
主要成果:
- 与独立的机器学习模型相比,CEEMDAN-Hybrid-Ensemble (CHES) 模型显著提高了预测性能,平均将R2的验证率提高了75%.
- 在多种水体 (河流和湖泊) 中,在多个时间分辨率 (每小时,每天,每两周) 中实现了高预测准确性.
- 证明了稳定的多步预测能力,具有高验证R2和低根平均平方误差 (RMSE).
结论:
- 将CEEMDAN与一组机器学习模型集成为准确预测藻类繁殖提供了一种强大的方法.
- 开发的CHES模型为环境监测和预警系统提供了一个强大的和多功能工具.
- 这项研究强调了集体方法在捕捉复杂的环境动态方面的显著好处.
相关概念视频
01:28Overview of Algae
220
The kingdom Archaeplastida encompasses red and green algae, along with land plants. Unlike other protists with chloroplasts that arose through secondary endosymbiosis, only red and green algae originated from primary endosymbiotic events. This diverse group of eukaryotic organisms contains chlorophyll and performs oxygenic photosynthesis.Algae exist in various forms, from large brown kelp in coastal waters to green scum in puddles and stains on rocks or soil. Some species are responsible for...
220
01:19Other Algae
96
The group Stramenopiles include some phototrophic microorganisms. Members of this group possess flagella covered in numerous short, hairlike extensions, a feature that inspired the group's name, derived from the Latin words for "straw" and "hair." Some of the main categories of Stramenopiles include diatoms, golden algae, and brown algae.Diatoms are unicellular, photosynthetic eukaryotes, with over 200 known genera. They play a key role in the planktonic communities of both marine and...
96
01:30Classification of Signals
935
In signal processing, signals are classified based on various characteristics: continuous-time versus discrete-time, periodic versus aperiodic, analog versus digital, and causal versus noncausal. Each category highlights distinct properties crucial for understanding and manipulating signals.
A continuous-time signal holds a value at every instant in time, representing information seamlessly. In contrast, a discrete-time signal holds values only at specific moments, often denoted as x(n), where...
A continuous-time signal holds a value at every instant in time, representing information seamlessly. In contrast, a discrete-time signal holds values only at specific moments, often denoted as x(n), where...
935
01:23Red Algae
155
Red algae, also known as rhodophytes, are primarily found in marine environments, though some species inhabit freshwater and terrestrial ecosystems. These organisms exist in both unicellular and multicellular forms, with some multicellular varieties reaching macroscopic sizes.As phototrophic organisms, red algae contain chlorophyll a; however, their chloroplasts lack chlorophyll b. Instead, they possess phycobiliproteins, which serve as major light-harvesting pigments, similar to those found in...
155