River flow forecasting by comparative analysis of multiple input and multiple output models form using ANN

Abstract ANN was used to create a storage-based concurrent flow forecasting model. River parameters in an unsteady must be modeled using model formulation based on learning storage change variable and instantaneous rate change. Multiple input-multiple output (MIMO) multiple input-single (MISO models three variants were anticipate rates the Tar Basin United States. Gamma memory neural networks, as well MLP TDNNs models, are this study. When issuing forecast, variables for river considered, which is why study includes them. While considering mass balance flow, proposed can provide real-time forecasting. Results obtained validated various statistical criteria such RMS error coefficient of correlation. For correlation value more than 0.96 indicates good results. results show fluctuations corresponding expressly implicitly provided variations.