Application of advanced learning methods for detecting network configuration in a smart water distribution system

Water distribution is one of the main pillars of modern society and accounts for a constant need of innovative solutions to age-old problems. There are many challenges associated to the aging infrastructure in large metropolitan areas as well as efficient operation of control structures, requiring improved decision support systems and autonomous operation based on available data. An extension of traditional methods with modern concepts is required such as using learning algorithms, smart meters and reactive programming for improving the quality of service in water distribution systems. This paper extends an IoT-based model with Deep Learning and automated test scenarios, while showing the effective application and comparison of learning techniques on experimental data in this domain. The experimental model is described from the hardware level to the IoT platform in a modern approach using the current state of software development and architectures for real-time data management.

[1]  Maria Pantoja,et al.  Introduction to Deep Learning , 2018, Deep Learning.

[2]  Ciprian Lupu,et al.  Modeling the Effects of Leaks on Measured Parameters in a Water Distribution System , 2017, 2017 21st International Conference on Control Systems and Computer Science (CSCS).

[3]  Lawrence O. Kehinde,et al.  Development of an MQTT-based IoT Architecture for Energy-Efficient and Low-Cost Applications , 2019 .

[4]  Luis M. Camarinha-Matos.  Application of Machine Learning in Water Distribution Networks : An Initial Study , 2007 .

[5]  Zheng Yi Wu,et al.  Applications of Deep Learning for Smart Water Networks , 2015 .

[6]  Zhen Wang,et al.  Designing of IoT Platform Based on Functional Reactive Pattern , 2018 .

[7]  Ji Feng,et al.  Deep Forest: Towards An Alternative to Deep Neural Networks , 2017, IJCAI.

[8]  Saso Dzeroski,et al.  Ensembles of Multi-Objective Decision Trees , 2007, ECML.

[9]  Luis M. Camarinha-Matos,et al.  Application of Machine Learning in Water Distribution Networks , 1998, Intell. Data Anal..

[10]  Ciprian Lupu,et al.  Real time implementation of IoT structure for pumping stations in a water distribution system , 2017, 2017 21st International Conference on System Theory, Control and Computing (ICSTCC).

[11]  Sandro Skansi,et al.  Introduction to Deep Learning , 2018, Undergraduate Topics in Computer Science.

[12]  Henrik Linusson,et al.  MULTI-OUTPUT RANDOM FORESTS , 2013 .

[13]  Bernardete Ribeiro,et al.  Introduction to Deep Learning Business Applications for Developers , 2018, Apress.

[14]  Ciprian Lupu,et al.  A modern approach for leak detection in water distribution systems , 2018, 2018 22nd International Conference on System Theory, Control and Computing (ICSTCC).

[15]  Vicenç Puig,et al.  Leak Localization in Water Distribution Networks using Deep Learning , 2019, 2019 6th International Conference on Control, Decision and Information Technologies (CoDIT).

[16]  Selvakumar Manickam,et al.  A Brief Review of Messaging Protocol Standards for Internet of Things (IoT) , 2019, J. Cyber Secur. Mobil..

[17]  Zheng Yi Wu,et al.  Optimized Deep Learning Framework for Water Distribution Data-Driven Modeling , 2017 .