Method for predicting gust during typhoon

Abstract

The invention discloses a method for predicting a gust during a typhoon, and the method is characterized in that the method comprises the following steps: setting a training set and a prediction set,and carrying out the normalization preprocessing of the training set and the prediction set; building a fuzzy training set and a fuzzy prediction set; initializing a population size N, the maximum number of iteration times, algorithm termination conditions, a smell concentration discrimination function, and a limiting parameter theta in a fruit fly optimization algorithm, and setting the value range of a penalty factor of a fuzzy support vector machine and the value range of a kernel parameter g; optimizing the penalty factor and the kernel parameter of the fuzzy support vector machine, obtaining the penalty factor and the kernel parameter of the fuzzy support vector machine after optimization, and obtaining the optimized fuzzy support vector machine; carrying out the fitting training of the fuzzy training set through the optimized fuzzy support vector machine, carrying out the prediction of the fuzzy prediction set, and achieving the prediction of the gust of the typhoon. The method is advantageous in that the prediction precision is higher and the prediction result is more effective and reliable.

Description

technical field [0001] The invention relates to a meteorological data prediction method, in particular to a method for predicting gusts during a typhoon period. Background technique [0002] Typhoon disasters are the most important natural disasters in the southeastern coastal areas of my country. They are usually accompanied by violent storms, which in turn cause secondary disasters such as floods, landslides, and mudslides, which bring great harm to the safety of people's lives and property, environmental protection, and economic development. Typhoons are characterized by strong catastrophe, destructiveness and suddenness. Timely grasp of typhoon information is very important for disaster relief. Due to the many factors affecting typhoon wind speed, the traditional wind speed prediction based on manual experience has large errors. Therefore, how to effectively and accurately predict typhoon wind speed has become the research content of many scholars and experts. At presen...

AI Technical Summary

Problems solved by technology

The random time series method uses ordered random data for analysis and processing, and the prediction results will vary greatly with the model order; Kalman filtering has the advantage of dynamically modifying weights, but it is difficult to establish state equations and prediction equations ; The artificial neural network has good robustness and generalization ability, but there are problems of unstable learning and memory, slow convergence speed and easy to fall into local minima; the support vector machine method can learn from small sample data and obtain Good generalization ability, after introducing the kernel function, effectively solves the disaster of dimensionality, but the traditional support vector machine cannot handle noise and fuzzy information

Images