ANN for detection and prediction of membrane fouling in water-treatment plants******

用于检测和预测水处理厂膜污染的 ANN******

• 批准号: 536518-2018
• 负责人: Dimopoulos, Nikitas
• 金额: $ 1.82万
• 依托单位: University of Victoria
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2018
• 资助国家: 加拿大
• 起止时间: 2018-01-01 至 2019-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=649314
• 关键词: ANN; detection; prediction; membrane; fouling

We propose to develop methods of detecting and predicting membrane-fouling in water treatment plants. **To achieve our goal, we propose to employ ensemble Neural Networks which incorporate post training regularization and hence can generalize well when trained using a sparse set of exemplars.**Our approach will complement and enhance methodologies developed by Pani Energy and will contribute to making the operation of such plants more energy efficient and less costly. This in turn will ensure the viability of the operation of such plants in under-developed areas of the world where pollution and drought are endemic. **

我们建议开发检测和预测水处理厂膜污染的方法。** 为了实现我们的目标，我们建议采用集成神经网络，它包含训练后的正则化，因此在使用稀疏样本集进行训练时可以很好地泛化。我们的方法将补充和加强帕尼能源开发的方法，并将有助于使这些工厂的运营更节能，成本更低。这反过来又将确保在世界上污染和干旱流行的欠发达地区运营这些工厂的可行性。**