Introduction

Underground coal fires pose significant environmental and health risks due to releasing CO₂ emissions. Predicting surface CO₂ flux accurately in underground coal fire areas is crucial for understanding the distribution of spontaneous combustion zones and developing effective mitigation strategies. In recent years, advanced machine learning techniques have shown promise in various carbon-related studies. This research uses an experimental approach to explore the power of advanced machine learning schemes for predicting CO₂ flux in underground coal fire areas.

Objectives

By leveraging the power of advanced machine learning schemes and experimental approaches, this research aims to provide valuable insights into CO₂ flux prediction in coal fire areas and inform environmental monitoring and management strategies.

Methods

The study involves the collection of an experimental dataset specific to underground coal fire areas, encompassing various parameters related to CO₂ flux and underground coal fire characteristics. Innovative feature engineering techniques are applied to capture the unique characteristics of underground coal fire areas and their impact on CO₂ flux. Different machine learning algorithms, including Natural gradient boosting regression (NGRB), Extreme gradient boosting (XGboost), Light gradient boosting (LGRB), and random forest (RF), are evaluated and compared for their predictive capabilities. The models are trained, optimized, and assessed using appropriate performance metrics.

Results

The NGRB model yields the best predictive performances with R² of 0.967 and MAE of 0.234. The novel contributions of this study include the development of accurate prediction models tailored to underground coal fire areas, shedding light on the underlying factors driving CO₂ flux. The findings have practical implications for delineating the spontaneous combustion zone and mitigating CO₂ emissions from underground coal fires, contributing to global efforts in combating climate change.

中文翻译：

---

用于预测地下煤火区基于 CO2 通量的实验方法的高级机器学习方案

 介绍

由于释放 CO₂ 排放，地下煤火会带来重大的环境和健康风险。准确预测地下燃煤区的地表 CO₂ 通量对于了解自燃区的分布和制定有效的缓解策略至关重要。近年来，先进的机器学习技术在各种碳相关研究中显示出前景。本研究使用一种实验方法来探索高级机器学习方案在预测地下煤火区域 CO₂ 通量方面的能力。

 目标

通过利用先进的机器学习方案和实验方法的力量，本研究旨在为燃煤地区的 CO₂ 通量预测提供有价值的见解，并为环境监测和管理策略提供信息。

 方法

该研究涉及收集特定于地下燃煤区域的实验数据集，包括与 CO₂ 通量和地下燃煤特性相关的各种参数。应用创新的特征工程技术来捕捉地下燃煤区域的独特特征及其对 CO₂ 通量的影响。评估和比较了不同的机器学习算法，包括自然梯度提升回归 （NGRB）、极端梯度提升 （XGboost）、轻度梯度提升 （LGRB） 和随机森林 （RF），以了解其预测能力。这些模型使用适当的性能指标进行训练、优化和评估。

 结果

NGRB 模型产生最佳预测性能，R² 为 0.967，MAE 为 0.234。这项研究的新贡献包括开发针对地下煤火区域的准确预测模型，阐明了驱动 CO₂ 通量的潜在因素。这些发现对于划定自燃区和减少地下煤火的 CO₂ 排放具有实际意义，有助于全球应对气候变化的努力。