Home

Welcome to watex wiki

Definition

WATex is an exploration open-source software using AI learning methods for water exploration. It is entirely written in Python programming language. It used the machine learning methods (unsupervised, supervised, dimensional reduction, ...) combine with geophysical methods for groundwater exploration. The objective is to bring a piece solution in a wide program of WATER4ALL in Africa and to participate in Sustainable Development Goals N6 achievement. The software has already made proof and was implemented in the Bagoue region in the north part of Cote d'Ivoire with a great success in flow rate prediction using the electrical methods such as Electrical resistivity profiling (ERP) and vertical electrical sounding (VES).

Problem solved

Bagoue region is known as an area strongly affected by climate change and faces a considerably drinking water shortage. According to the area regional hydraulics report from the Ivorian Hydraulic Ministry, 40.98% of boreholes are unsuccessful after drilling, 33.25 % are unsustainable during the dry seasons and 25.77% of the water of the productive boreholes dried up after three years of use. The average FR observed in this area fluctuates between 1 and 3 m3/h. Therefore, WATex is used to predict the flow rate (FR) using the SVMs in this context so the inferred relationship can be applied to any dataset that is consistent for future campaigns of drinking water supply (CDWS) or a local groundwater exploration. The methods used here were ERP and VES. Indeed, the ERP and VES are cheap geophysical subsurface imaging methods. They are most preferred to find groundwater during the CDWS, especially in developing countries. However, despite the use of both methods, the numerous unsuccessful drillings, due to their wrong locations, have considerably increased the budget of the project and thereby limiting the number of boreholes previously intended for the population. To solve this problem, a technique was developed using one famous method of artificial intelligence called support vector machines to predict the FR before the drilling operations. To check the efficiency of the proposed approach, the technique was tested with the data from a region in the northern part of Cote d’Ivoire (West Africa), which faced a considerable water shortage. The results show 77% capability to predict an accurate FR and 83% when the problem is addressed to the population living in a rural area. Henceforth, the proposed technique implemented via WATex can be used to select the right locations expecting to give the recommended FR to minimize the rate of unsuccessful drillings, and indirectly reduce the problem of water scarcity.

Development progress

Currently, WATex works with the learning methods enumerated below:

• Support vector machines
• Neighbors: KNN
• Trees: DTC
• Ensemble methods (RandomForests, Bagging and Pasting, Boosting)
• Artificial neural networks ANN
• Apriori
• Kernel Principal Component Analysis k-PCA
• t-distributed Stochastic Neighbor Embedding t-SNE
• Randomized PCA
• Locally Linear Embedding (LLE)

and implements the geophysical methods below:

• Electrical Resistivity Profiling
• Vertical Electrical Sounding
• Countrolled Source Audio-frequency Magnetotelluric

and the other methods of development are still ongoing. Other AI algorithms implemented will be added as things progress. Furthermore, WATex intends to add to its functionalities the water sanitation aspect for population welfare by bringing a piece of the solution to their daily problems. The latter goal should not be developed for the first release.

• Note: The package organization above is not updated. It was the first structured package so some modules have been, deprecated, renamed and many other new functionalites were added. Moreover, refer to WATex Objectives for additionals details.