Waterlogging model prediction method based on firefly algorithm

Abstract

The invention belongs to the technical field of waterlogging models, and particularly relates to an waterlogging model prediction method based on a firefly algorithm, and the method comprises the steps: 1, training a weight and a threshold of a neural network through employing a training error of the neural network as a fitness function of a firefly individual, and obtaining an optimal parameter;2, constructing a waterlogging prediction model by using the optimal parameters obtained in the step 1; according to the firefly algorithm-based waterlogging prediction model provided by the invention, training the weight and the threshold of the neural network by adopting the firefly swarm algorithm, and performing waterlogging prediction by adopting the trained neural network model. According tothe method, the defects that a traditional rainfall ponding model needs a large amount of basic data information, requires elaborate hydrological and physical processes and the like are overcome, meanwhile, the defect that a neural network is prone to falling into local minimum is overcome, and the method has important significance in accurate prediction of urban inland inundation and acceleration of the process of a smart city.

Description

Technical field [0001] The invention belongs to the technical field of waterlogging models, and specifically relates to a waterlogging model prediction method based on a firefly algorithm. Background technique [0002] With the development of society and the acceleration of the process of industrialization, the ecological environment has been destroyed, and waterlogging disasters occur frequently. Waterlogging will not only cause traffic paralysis, but also cause great losses to the people's personal safety and property. The research and prediction of waterlogging models are particularly important. Predicting urban waterlogging in advance and improving the accuracy of prediction can effectively reduce the impact of urban waterlogging and accelerate the process of smart cities. [0003] The traditional rainfall water model SWMM requires a lot of basic data and requires proficiency in hydrophysical processes, which limits the accuracy and use of the model. With the development of s...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to overcome the deficiencies of the existing waterlogging model and solve the shortcoming that the neural network is easy to fall into the local minimum. The present invention provides a waterlogging model prediction method based on the firefly algorithm

Images