USE OF MACHINE LEARNING IN GROUNDWATER LEVEL FORECASTING
Journal: International Journal of Advances in Agricultural Science and Technology (IJAAST) (Vol.7, No. 6)Publication Date: 2020-06-30
Authors : Mukesh Kumar Mehla; Mahesh Kothari; Harsh Upadhyay; Priyanka Jagtap;
Page : 19-24
Keywords : Groundwater Level; Machine Learning; ANN; SVM and Soft computing;
Abstract
Knowledge about the current groundwater levels of an area play a very important role in proper utilization and management of groundwater supply. Use of machine learning for groundwater levels prediction can drastically change the way forecasting, monitoring and management of groundwater is carried out traditionally, reducing the need of costly surveys and reduce dependence on manual monitoring of wells. Artificial Neural Networks (ANNs) have been used progressively in recent years for various hydrological applications because of their ability to truthfully model the complicated non-linear relationships. Availability of proper database and frequency of data collection for creating accurate and usable dataset is one of many constraints faced when using machine learning algorithms for groundwater level prediction. Use of ANN, GP, SVM and ELM for groundwater levels prediction has many advantages over traditional methods; Performance of SVM can be enhanced by using SVM hybrid models such as SVM-QPSO and SVM-RBF. Adaptive neural fuzzy inference system (ANFIS) and genetic programming (GP) are the artificial intelligence tools that help to predict and simulate groundwater levels. GP simulated equations decrease the computational effort by using common simulation packages that can yield results with acceptable accuracy. Statistical indicators RMSE, r, R2, MAE and MAPE can be used for the comparison of these models and find best suited model under a given condition.
Other Latest Articles
- Normalization process of cardiac operations in COVID-19 pandemic
- Livelihood Security through Tea Plantation in Tribal Area of Chhattisgarh
- The dynamics of language game involving address terms
- Photographs as methodological strategies: searching for Brazilian, Mozambican and Angolan addressing pronominal forms
- The social-professional classification: a way to approach the study of forms of address
Last modified: 2020-06-14 17:50:40