Development of an artificial neural network for identification of used button cell batteries

开发用于识别废旧纽扣电池的人工神经网络

• 批准号: 516127-2017
• 负责人: Fraser, Andrew
• 金额: $ 1.82万
• 依托单位: Conestoga College Institute of Technology & Advanced Learning
• 依托单位国家: 加拿大
• 项目类别: Applied Research and Development Grants - Level 1
• 财政年份: 2017
• 资助国家: 加拿大
• 起止时间: 2017-01-01 至 2018-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=641266
• 关键词: Development; artificial; neural; network; identification

Alkaline batteries in all shapes and sizes are used in many applications and comprise 78% of all batteries soldinto the Canadian market every year. Alkaline (non-rechargeable) and Lithium-Ion (rechargeable) batteriescontain Zinc and rare Earth metals such as Lithium, Cadmium, and Nickel. If these batteries are recycledresponsibly, they will provide new sources for fertilizing agricultural fields, save energy and resources, protectthe environment, and divert waste from landfills.However, considering the volume and variety of waste batteries, automated systems require the ability to sortby chemistry in order to ensure proper recycling. Sorting results in better control over environmental outcomesas separated battery types can be processed by the best specialty recyclers using technology designed forrecycling of specific battery chemistries.In this feasibility study project, we will investigate different ways to employ new technologies for improvingthe efficiency and productivity of household batteries recycling. The core of our study will be to develop areliable method for identifying different types of used button cell batteries using machine vision and ArtificialNeural Networks (ANNs).

各种形状和尺寸的碱性电池在许多应用中使用，每年占加拿大市场上销售的所有电池的78%。碱性（不可充电）和锂离子（可充电）电池含有锌和稀土金属，如锂、镉和镍。如果这些电池被负责任地回收利用，它们将为农田施肥提供新的来源，节省能源和资源，保护环境，并从垃圾填埋场转移废物。然而，考虑到废电池的数量和种类，自动化系统需要能够对化学物质进行分类，以确保适当的回收。分类可以更好地控制环境后果，因为分离的电池类型可以由最好的专业回收商处理，使用专为回收特定电池化学物质而设计的技术。在这个可行性研究项目中，我们将研究不同的方法来采用新技术来提高家用电池回收的效率和生产力。我们研究的核心将是利用机器视觉和人工神经网络（ANNs）开发可靠的方法来识别不同类型的废旧纽扣电池。