Development of neural networks in predicting Caspian sea tides based on Levenberg Marquardt algorithm

The purpose of this research is to show the use of neural networks in predicting the tides of the Caspian Sea. Forecasting the current level of seawater is very useful in the shipping industry and is one of the most important parameters in marine geodesy, oceanography, and geophysics. Sea level has been done in different ways. Among the factors that affect the instantaneous sea level changes, the tidal factor has been investigated in this research. Nowadays, artificial intelligence methods are used to predict tides, these methods can fill information gaps the above items are among the advantages of intelligent methods in analyzing and predicting tides. Accordingly, this study method has been used in the preparation of this research. The nature of this research is quantitative and applied, and a descriptive research method has been used. In this research, one of the types of artificial intelligence has been worked with, which is neural networks. Neural networks have been used to validate the primary data and predict the Caspian Sea tides in this research. According to the findings obtained by Geo Tide software and neural networks, the water level of the Caspian Sea is decreasing because is clear in this research. It was found that the neural networks are more effective for predicting the tides, the water level of the Caspian Sea will reach -28 meters from the open water level in the future.