Projection Pursuit Regression Model Based on Real-coded Genetic Algorithm for Flood Forecastion

Combining the advantages of genetic algorithm (GA) and projection pursuit regression (PPR), this article firstly uses improved Hermit polynomial as the ridge function of projection pursuit regression model. And then adopted the real-coded genetic algorithm to optimize the projection direction, a forecasting model for peak flow of short flood forecasting is presented. Applied the presented model to forecasting the flood of the Wujiang River at Wulong station, and compared with the BP neural network method. Computing results show that, the presented model has a strong advantage of dimensionality reduction adaptability than BP network in dealing with the poor fitting data of one dimension space, the forecasting accuracy is improved, and can be applied in hydrological simulation and forecasting.