Predictive modeling framework accelerated by GPU computing for smart water grid data-driven analysis in near real-time. (November 2022)

• Record Type: Journal Article
• Authors: Kalfarisi, Rony; Chew, Alvin; Cai, Jianping; Xue, Meng; Pok, Jocelyn; Wu, Zheng Yi
• Is Part Of: Advances in engineering software. Volume 173(2022)
• Publication Date: 2022-11
• Subjects: Predictive modelling -- Smart water grid -- Data-driven -- Statistical methods -- Machine learning -- Deep learning -- Near real-time; Computer-aided engineering -- Periodicals; Engineering -- Computer programs -- Periodicals; Engineering -- Software -- Periodicals; Periodicals; 620.0028553
• DOI: 10.1016/j.advengsoft.2022.103287
• Languages: English
• ISSNs: 0965-9978

Abstract

Highlights: Developed predictive modeling framework with data preprocessing, statistical methods, machine/deep learning algorithms. Flexible and accelerated by the latest graphics processing unit computing technology. Addressed challenges in developing accurate and robust model for near real-time applications. Conducted impact analysis of training data size, model update frequency, and input data size on model performance. Achieved the prediction accuracy of 91% and 98% for flows and pressures, respectively using real monitoring data. Abstract: With the increase adoption of monitoring technology for Smart Water Grid (SWG) system, accurate prediction of SWG status is essential for water companies to effectively operate and manage water networks. Although different data-driven predictive techniques have been developed over last two decades with various degree of success, predictive modeling is not widely adopted in practice. The challenges remain in (1) developing accurate and robust model for near real-time applications; (2) the selection of training data size, model update frequency, and input data size for competent model performance. Therefore, in this paper, a versatile framework is developed by integrating data preprocessing procedures with various statistical methods, machine learning, and deep learning algorithms. It is flexible and accelerated by the latest graphics processing unit computing technology. The case study using the real-world monitoring data shows that the prediction accuracy of 91% and 98% has been achieved for flow and pressures, respectively.