Kouadio, Kouao Laurent, Kouame, Loukou Nicolas, Drissa, Coulibaly, Mi, Binbin, Kouamelan, Kouamelan Serge, Gnoleba, Serge Pacôme Déguine, Zhang, Hongyu, and Xia, Jianghai
Unsuccessful drillings are issues in groundwater exploration using electrical resistivity profiling (ERP) and vertical electrical sounding (VES). Many geophysical companies spend a lot of money without obtaining the flow rate (FR) required during the campaigns for drinking water supply (CDWS). To solve this problem, we applied the support vector machines (SVMs) to real‐world data to predict the FRs before any drilling operations. First, from the ERP and VES, the features such as shape, type, power, magnitude, pseudo‐fracturing index, and ohmic‐area were defined including the geology of the survey area. Second, the FRs were categorized into four classes (dry: FR0 (FR = 0), unsustainable: FR1 (0 < FR ≤ 1), and productive boreholes: FR2 (1 < FR ≤ 3) and FR3 (FR > 3 m3/hr)) and associated with the features to compose two separated data sets: a multiclass data set (D $\mathcal{D}$) for common prediction during the CDWS and a binary data set Db ${\mathcal{D}}_{b}$ (FR < FR2, FR ≥ FR2) addressed to the population living in a rural area. Features were vectorized and data were transformed before feeding to the SVM algorithms. As a result, the SVM models performed 77% of good predictions on D $\mathcal{D}$ and 83% on Db ${\mathcal{D}}_{b}$. Better performances with the optimal hyper‐parameters in D $\mathcal{D}$ (81.61%) and Db ${\mathcal{D}}_{b}$ (87.36%) were achieved using the polynomial and radial basis function kernels respectively. Furthermore, the learning curves have shown that the performance scores on D $\mathcal{D}$ can be improved if larger training data becomes available (275 test samples at least) while it is not necessarily so for Db ${\mathcal{D}}_{b}$. As a benefit, the proposed approach could minimize the rate of unsuccessful drillings during future CDWS. Plain Language Summary: The electrical resistivity profiling and the vertical electrical sounding are cheap geophysical subsurface imaging methods. They are most preferred to find groundwater during the campaigns of drinking water supply, 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 flow rate (FR) before the drilling operations. To check the efficiency of the proposed approach, the technique was tested with 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 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. Key Points: The support vector machines (SVMs) are used to predict groundwater flow rate (FR) before the drilling operationsPrediction data are gathered from electrical profiling, vertical electrical sounding, and boreholesThe SVM models correctly perform 77% and 83% of FR predictions for multiclass and binary data sets respectively [ABSTRACT FROM AUTHOR]