Rainfall Estimation using PL-HONN Model

M. Zhang (USA)


Multi-Polynomial, Logarithm, High Order NeuralNetwork, Rainfall, Simulation


Multi-Polynomial with Logarithm function High Order Neural Network (PL-HONN) model has been developed in this paper. The PL-HONN model for estimating heavy rainfall from satellite data has been tested as well. The PL-HONN model has 1.8% to 5.1% more accuracy than M-PHONN (Multi-Polynomial High Order Neural Network model) and PHONN (Polynomial Higher Order Neural Network) Models. Using ANSER expert system, the average rainfall estimate errors for the total precipitation event could be reduced to less than 20%.

Important Links:

Go Back