Comparison of Soft Computing Techniques for Trap Efficiency- A Reservoir Case Study

  • Qamar Sultana Associate Professor, Civil Engineering Dept., Muffakham Jah College of Engg & Tech., Hyderabad, 500034, Telangana, India.
  • M. Gopal Naik Professor, Civil Engineering Dept., University College of Engineering(UCE), Osmania University, Hyderabad, Telangana, India
Keywords: Trap Efficiency, Artificial Neural Networks, Fuzzy Logic.


Reservoirs are built across the streams to store water and serve the human kind during the periods of requirement. But along the water, the sediment also gets stored in the space created and the quantity of water stored in the reservoir gets reduced. There are different methods to determine the quantity of this sediment. Of all the methods, the simplest way is through the concept of trap efficiency of reservoir. Different empirical formulae are available to determine the trap efficiency of reservoir. In this study, the soft computing methods such as Artificial Neural Networks (ANN) and Fuzzy Logic are applied using the Matlab software has been. These techniques are applied for the Sriramsagar reservoir of Godavari basin. The root mean square error (RMSE) value for the trap efficiency values generated by the ANN and the fuzzy logic indicates that the ANN model simulates the trap efficiency values with less error, when compared to that of the fuzzy logic model values. The model efficiency value for the ANN model is positive, where as it is negative with very less value for the fuzzy logic model, which indicates that the efficiency of the ANN model is better than that of fuzzy logic model.