A Smart Drought Forecasting Method Combining Distribution Estimation Algorithm and Extreme Learning Machine

Abstract

The present invention discloses a drought intelligent forecasting method that integrates distribution estimation algorithm and extreme learning machine, proposes a method for successfully integrating distribution estimation algorithm (EDA) and extreme learning machine (ELM), and constructs an EDA-ELM hybrid model; according to research According to the characteristics of the region, select and calculate the corresponding drought index to represent the drought, and output it as an EDA-ELM model; collect drought-causing factors related to the cause of drought to build a large data set, and use information theory to screen out the key drought-causing factors that are most closely related to the cause of drought Factors, as model input; set the parameters in the EDA-ELM hybrid model, debug the model structure, and apply it to drought prediction. This method integrates relevant knowledge in the fields of informatics, statistics, and hydrometeorology. It has the advantages of high prediction accuracy, strong generalization ability, and wide application range. It provides an effective path for drought prediction based on big data.

Description

technical field [0001] The invention belongs to the fields of the Internet and big data, and specifically relates to a drought intelligent prediction method that integrates a distribution estimation algorithm and an extreme learning machine. Background technique [0002] Drought is a type of meteorological disaster that causes the most serious economic loss in the world among many natural disasters. my country has a vast territory, and the drought disaster is particularly severe, which has become an important factor restricting the sustainable development of my country's society and economy. Strengthening the research on mid- and long-term drought prediction and early warning, and improving the forecast period and prediction accuracy of drought will not only provide the necessary technical support for my country's drought relief and disaster reduction work, but also ensure my country's food security, water supply security, ecological security, and sustainable social and econ...

AI Technical Summary

Problems solved by technology

However, no one has successfully integrated the distribution estimation algorithm and the extreme learning machine model, which is a gap in the improvement of the model.

Images