Prediction of groundwater level in Urmia Plain aquifer using hybrid model of wavelet Transform-Extreme Learning Machine based on quantum particle swarm optimization

Today, due to the importance of sustainable groundwater management, groundwater level modeling and forecasting are used to assess and evaluate water resources. The purpose of this study is to evaluate the performance of two models of Extreme Learning Machines (ELM) and Artificial Neural Network (ANN) and the combination of two models with wavelet transmission algorithms (W-ELM and W-ANN), which ultimately to increases the predictive power and optimization of input weights (the weights between the input and hidden layers) of models, Quantum Particle Swarm Optimization algorithm (QPSO) has been used. Also, in this study, the data of Ground Water Level of observation wells (GWL), precipitation (P) and average temperature (T) of Urmia Plain aquifer with a time series of 36 years (1981 – 2017) which were collected on monthly scale, are used. Also, in order to evaluate the performance of models, correlation coefficient (R), Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) were used. In this regard, 80% of the data (September 1981 to August 2010) are used for training section and 20% of data (September 2010 to August 2017) used for the test section of models. Based on the results of this study, the hybrid model of W-ELM-QPSO with correlation coefficient (R) 0.991, 0.983 and 0.975, respectively for periods of one, two and three months in the test section, have a better performance than other models and also in addition to predicting power, this model has a high speed in terms of training and testing speed than other models.