返回
Adaptive greedy learning-optimized Gaussian mixture models for streamflow stochastic simulation
DOI:10.1016/j.ecmx.2026.102125.png)
摘要
En 中文
• 提出了一种用于径流模拟的贪心学习优化的GMM。
• 使用数据引导的候选分量生成和两阶段参数精炼。
• 相比EM训练,实现了更高的对数似然值和更简洁的混合模型。
• 更好地保留了统计特性、时间依赖性、尾部特征、干旱行为和不确定性。
Keyword:
Streamflow stochastic simulation
Gaussian mixture model
Greedy learning
Adaptive structure determination
Temporal dependence modeling
GL
Greedy Learning
GMM
Gaussian Mixture Model
GL-GMM
Greedy Learning-optimized Gaussian Mixture Model
EM
Expectation Maximization
ARMA
Autoregressive Moving Average model
ARIMA
Autoregressive Integrated Moving Average model
SAR
Seasonal Autoregressive model
GAN
Generative Adversarial Network
SCG
Splitting-based Candidate Generation
CDF
cumulative density function
RMSE
Root Mean Square Error
MAE
Mean Absolute Error
MAPE
Mean Absolute Percentage Error
emp
Empirical
the
Theoretical
SD
Standard Deviation
CS
Skewness Coefficient
RS
Autocorrelation Coefficient for Scattered Time Steps
ACF
Autocorrelation coefficient
Lag-1 ACF
Lag-1 Autocorrelation Coefficient
CR
Coverage Rate
IW
Interval Width
AIC
Akaike Information Criterion
CLs
Confidence Levels
AI总结
对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
E
IF:
7.6
论文数:
1.3K
被引数:
4.5K
机构
引用论文
Soft sensor model development in multiphase/multimode processes based on Gaussian mixture regression基于高斯混合回归的多相/多模过程软测量模型开发
Data-driven scenario generation of renewable energy production based on controllable generative adversarial networks with interpretability基于可解释性可控生成对抗网络的数据驱动可再生能源生产场景生成
APPLIED ENERGY
IF11
Optimized Two-Stage Planning Model for Integrating Compressed Air Energy Storage With Uncertain Correlated Wind Farms in Power Systems电力系统中压缩空气储能与不确定相关风电场集成的优化两阶段规划模型
High-dimensional scenario generation method joint-driven by multiple correlations for hydro-wind-photovoltaic基于多重相关性联合驱动的水电-风电-光伏高维场景生成方法
Applied Energy
IF11

