Machine learning approach to monitoring and decarbonizing mineral processes

用于监测和脱碳矿物过程的机器学习方法

• 批准号: 570866-2021
• 负责人: Sari, YukselAsli
• 金额: $ 1.46万
• 依托单位: Queen's University
• 依托单位国家: 加拿大
• 项目类别: Alliance Grants
• 财政年份: 2021
• 资助国家: 加拿大
• 起止时间: 2021-01-01 至 2022-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=719598
• 关键词: Machine; learning; approach; monitoring; decarbonizing

In Canada's 2030 Agenda for Sustainable Development, one of the goals is to take urgent action to combat climate change and its impacts. According to this agenda, the target is to reduce Canada's total greenhouse gas (GHG) emissions by 30%. In order to transition to renewable energy, solar and wind power in combination with energy storage such as lithium-ion batteries can be used to replace fossil fuels. Although this approach has a great potential in decreasing the carbon emissions in the long-term, it is also important to minimize the environmental effects of the production of these batteries. The proposed research aims to reduce the carbon emissions in mineral processing activities using machine learning. In this study, a novel approach to minimizing the carbon footprint of a mining-processing plant will be developed in three stages: (1) Performing feature engineering to determine the key parameters affecting the carbon emissions, (2) analyzing the historical data to develop models of estimation and (3) developing a real-time control approach using machine learning. The adaptable machine learning model will be used to minimize fuel consumption by processing plants. This will allow increasing the production of required minerals for lithium-ion batteries in Canada while meeting the GHG reduction targets. As the lithium nickel cobalt aluminum oxides (NCA) used in these batteries represent an important part of Canada's economy and considering their increasing demand by the electric automotive sector in the future, continuing the production with reduced emissions is critical for the economy and the environment.

在加拿大的《2030年可持续发展议程》中，目标之一是采取紧急行动应对气候变化及其影响。根据这一议程，目标是将加拿大的温室气体排放总量减少30%。为了过渡到可再生能源，太阳能和风能与锂离子电池等储能相结合，可以用来取代化石燃料。虽然这种方法在长期减少碳排放方面具有很大的潜力，但最大限度地减少这些电池生产对环境的影响也很重要。拟议的研究旨在使用机器学习减少矿物加工活动中的碳排放。在这项研究中，将开发一种新的方法来最大限度地减少采矿加工厂的碳足迹，分为三个阶段：（1）执行特征工程，以确定影响碳排放的关键参数，（2）分析历史数据，以开发估计模型和（3）使用机器学习开发实时控制方法。可适应的机器学习模型将用于最大限度地减少加工厂的燃料消耗。这将使加拿大锂离子电池所需矿物的产量增加，同时实现温室气体减排目标。由于这些电池中使用的锂镍钴铝氧化物（NCA）是加拿大经济的重要组成部分，考虑到未来电动汽车行业对它们的需求不断增加，继续生产并减少排放对经济和环境至关重要。