A Hybrid Precipitation Prediction Method based on Multicellular Gene Expression Programming

TL;DR

This paper introduces WTGEPRP, a hybrid precipitation forecasting method combining multicellular gene expression programming and wavelet analysis, improving accuracy over existing models for non-linear, noisy data.

Contribution

It proposes a novel hybrid algorithm that enhances precipitation prediction by integrating advanced data mining and denoising techniques, outperforming several existing models.

Findings

WTGEPRP shows superior fitting and forecasting accuracy.

The method effectively handles non-linear and noisy precipitation data.

It demonstrates good potential for practical meteorological applications.

Abstract

Prompt and accurate precipitation forecast is very important for development management of regional water resource, flood disaster prevention and people's daily activity and production plan; however, non-linear and nonstationary characteristics of precipitation data and noise seriously affect forecast accuracy. This paper combines multicellular gene expression programming with more powerful function mining ability and wavelet analysis with more powerful denoising and extracting data fine feature capability for precipitation forecast modeling, proposing to estimate meteorological precipitation with WTGEPRP algorithm. Comparative result for simulation experiment with actual precipitation data in Zhengzhou, Nanning and Melbourne in Australia indicated that: fitting and forecasting performance of WTGEPRP algorithm is better than the algorithm Multicellular Gene Expression Programming-based Hybrid Model for Precipitation Prediction Coupled with EMD, Supporting Vector Regression, BP Neural Network, Multicellular Gene Expression Programming and Gene Expression Programming, and has good application prospect.