A hierarchical reinforcement learning-based vehicle-to-grid dispatch architecture for car parks: Integrating proximal policy optimisation and a large language model within a dual-agent framework

Jinjia Cao

doi:10.1016/j.seta.2026.105158

Adaptive ensemble forecasting of industrial-scale biogas production under fluctuating operating conditions

2026-08-04

0

PRE

AI

J

Jiansheng Zhao

Y

Yuanda Cheng

X

Xiangxiang Jiao

X

Xinyi Li

J

Jing Hua *

DOI：10.1016/j.seta.2026.105281

Original

Original request for help

Save

Abstract

En 中文

• Dynamic LSTM-XGB ensemble handles industrial biogas fluctuations. • Interpretable ML revealed core operational drivers of biogas production. • SHAP analysis validates VFA as a key early warning metric for AD. • Adaptive framework reduces prediction errors by 23.6% over baselines. • Dynamic weighting paradigm generalizes to other fluctuating processes.

Keywords:

Industrial-scale biogas production

Anaerobic digestion

Dynamic ensemble forecasting

Data volatility

Machine learning

Adaptive ensemble forecasting of industrial-scale biogas production under fluctuating operating conditions

Abstract

Journal

Organization

Cited Papers

Citing Papers