Abstract:In recent years, the phase-space reconstruction method has usually been used for mid- and long-term runoff predictions. However, the traditional phase-space reconstruction method is still needs to be improved. Using the genetic algorithm to improve the phase-space reconstruction method, a new nonlinear model of monthly runoff is constructed. The new model does not rely heavily on embedding dimensions. Recognizing that the rainfall-runoff process is complex, affected by a number of factors, more variables (e.g. temperature and rainfall) are incorporated in the model. In order to detect the possible presence of chaos in the runoff dynamics, chaotic characteristics of the model are also analyzed, which shows the model can represent the nonlinear and chaotic characteristics of the runoff. The model is tested for its forecasting performance in four types of experiments using data from six hydrological stations on the Yellow River and the Yangtze River. Results show that the medium-and long-term runoff is satisfactorily forecasted at the hydrological stations. Not only is the forecasting trend accurate, but also the mean absolute percentage error is no more than 15%. Moreover, the forecast results of wet years and dry years are both good, which means that the improved model can overcome the traditional ''wet years and dry years predictability barrier,'' to some extent. The model forecasts for different regions are all good, showing the universality of the approach. Compared with selected conceptual and empirical methods, the model exhibits greater reliability and stability in the long-term runoff prediction. Our study provides a new thinking for research on the association between the monthly runoff and other hydrological factors, and also provides a new method for the prediction of the monthly runoff.

Application of Genetic Programming to Runoff Forecasting

GENETIC PROGRAMMING FOR POWER SYSTEM LOAD FORECASTING

Optimal operation of water supply systems based on genetic algorithm

A Genetic Programming Approach to System Identification of Rainfall-Runoff Models

A segregated genetic programming for bioprocess modelling with outliers

Genetic Programming Based On An Adaptive Regularization Method

Runoff Forecasting Using Machine-Learning Methods: Case Study in the Middle Reaches of Xijiang River

State-of-art of genetic programming applications in water-resources systems analysis

Short-term Runoff Prediction Optimization Method Based on BGRU-BP and BLSTM-BP Neural Networks

A Hydrological Process-Based Neural Network Model for Hourly Runoff Forecasting

Short-term forecasting of traffic flow based on genetic algorithm and bp neural network

Bioprocess modeling using genetic programming based on a double penalty strategy

Mid- and long-term runoff predictions by an improved phase-space reconstruction model

Interpretable probabilistic modeling method for runoff prediction: A case study in Yangtze River basin, China

GECA Proposed Ensemble–KNN Method for Improved Monthly Runoff Forecasting

Annual Runoff Forecasting Based on Multi-Model Information Fusion and Residual Error Correction in the Ganjiang River Basin

Application of Genetic Algorithm for Model Parameter Calibration in Daily Rainfall-Runoff Simulations with the Xinanjiang Model

A Novel Runoff Forecasting Model Based on the Decomposition-Integration-Prediction Framework

Application of improved seasonal GM(1,1) model based on HP filter for runoff prediction in Xiangjiang River

Exploring the use of multi-gene genetic programming in regional models for the simulation of monthly river runoff series

A Medium and Long-Term Runoff Forecast Method Based on Massive Meteorological Data and Machine Learning Algorithms