Paulo Adriano FERNANDES

Paper reading on water management by using machine learning.

Transcript

1. Application of machine learning in Water management
2. The water is of the major importance to all living things; Up to 60% of our body weight is water. Can we leave without it? Management of water is something very important for us now as well as the future generations. I am interested in learning how to manage water by using machine learning, the following slides are summarising some papers that read on water management by using machine learning.
3. Water quality prediction using machine learning methods In this paper the performance of artificial intelligence (AI) techniques with the artificial neural network (ANN), group method of data handling (GMDH) as well as the support vector machine (SVM) was investigated for predicting water quality components of Tireh River locate in the southwest of Iran. Different types of transfer and kernel functions were tested, respectively to develop the ANN and SVM. The results shows that ANN and SVM have good performance in predicting the components of water quality, but RBF (radial basis function) as transfer and kernel have the best performance among the tested functions. The results showed that the GMDH model has acceptable performance but its accuracy is slightly less than ANN and SVM in predicting the components of water quality. According to the error index, SVM was the most accurate model. Water quality prediction using machine learning methods | Water Quality Research Journal | IWA Publishing (iwaponline.com)
4. Application of machine learning techniques in water distribution networks assisted by domain experts The European Esprit Waternet is being done in the context of this project, where they applied machine learning techniques. To capture knowledge from historic information collected during the operation of a water distribution network, a learning system was used in this project. That captured knowledge was expected to contribute to improve the operation of the network. The ideas pictured in this paper describe the first development phase of this learning system, and focusing specially in the practical methodology adopted. It discussed also about the interaction between classes of human experts and the learning system, to conclude they come up with some preliminary experimental results. 978-1-4757-5547-3_Book_PrintPDF.pdf (springer.com)
5. Using Artificial Intelligence for Smart Water Management Systems This paper introduces the brief principles of artificial intelligence for water utilities embarking on this digital transformation to improve their water distribution operation in general, and how to address particular unaccounted problems in water network. The brief describes some of the most extended applications of big data analytics and artificial intelligence-related algorithms in water supply, discusses how water utilities can pilot artificial intelligence toward the prognosis of unaccounted-for-water, and presents recommendations for implementation and preliminary cost estimates. Using Artificial Intelligence for Smart Water Management Systems (ADB Brief No. 143)
6. Machine learning model and strategy for fast and accurate detection of leaks in water supply network This study developed Machine Learning (ML) models to detect leaks in the WDN. Water pressure data under leaking versus non-leaking conditions were generated with holistic WSN simulation code EPANET considering factors such as the fluctuating user demands, data noise, and the extent of leaks, etc. The results indicate that Artificial Neural Network (ANN), a supervised ML model, can accurately classify leaking versus non-leaking conditions; it, however, requires balanced dataset under both leaking and non-leaking conditions, which is difficult for a real WSN that mostly operate under normal service condition. Autoencoder neural network (AE), an unsupervised ML model, is further developed to detect leak with unbalanced data. The results show AE ML model achieved high accuracy when leaks occur in pipes inside the sensor monitoring area, while the accuracy is compromised otherwise. This observation will provide guidelines to deploy monitoring sensors to cover the desired monitoring area. The ML model and leak detection strategy can be readily deployed for in-service WSNs using data obtained with internet-of-things (IoTs) technologies such as smart meters. Machine learning model and strategy for fast and accurate detection of leaks in water supply network (springeropen.com)