A Forecast Heuristic of Back Propagation Neural Network and Particle Swarm Optimization for Annual Runoff Based on Sunspot Number

Water, Journal Year: 2024, Volume and Issue: 16(19), P. 2737 - 2737

Published: Sept. 26, 2024

Runoff prediction is of great importance to water utilization and water-project regulation. Although sun activity has been considered an important factor in runoff, little modeling constructed. This study put forward a forecast heuristic combining back propagation neural network (BPNN) particle swarm optimization (PSO) for annual runoff based on sunspot number applied it the Yellow River China period 1956–2016 assessed contribution by placing sole BPNN time series as contrast. First, made up calibration PSO optimization: (1) we use historical data calibrate models obtain training testing stages; (2) minimize difference between predicted both linear equation forecasting stage. Second, application offers interesting findings: while obtains first-class with ratio >85% <20% absolute error stages, can achieve similar performance stage; better years lower number; (3) besides influence activity, atmospheric circulation, usage, regulation do play roles measured or natural some extent. could provide useful insights into further under this world.

Language: Английский

A hydrogeological acrostic: in honour of Ghislain de Marsily

Comptes Rendus Géoscience, Journal Year: 2023, Volume and Issue: 355(S1), P. 1 - 8

Published: Nov. 24, 2023

Language: Английский