Urban CA model parameter optimization method based on biogeographic optimization algorithm

Abstract

The invention discloses an urban CA model parameter optimization method based on a biogeographic optimization algorithm, and the method comprises the steps: carrying out the rasterizing of an urban present land-use map, enabling each grid to be corresponding to one cell, and building a land-use matrix; extracting the urban land-use change and spatial variable data, carrying out spatial sampling, and obtaining a training sample; building the function relation between a spatial variable and urban expansion, and constructing a fitness function through employing the difference between a function fitting value and a true value; initializing parameters of a BBO algorithm, employing the BBO algorithm for training according to historical sample data, obtaining a variable weight, and calculating the urban development probability; building a transformational rule of a BBO-CA model, and calculating the final cell transformation probability of each cell according to the transformational rule of a non-urban cell; and carrying out urban expansion simulation based on a built BBO-CA model. The method constructs a biogeographic optimization cell automatic machine model, enables the model to be used in urban expansion simulation, and facilitates more true and accurate simulation of urban development.

Description

technical field [0001] The invention belongs to the technical field of metrological geography, and relates to a method for optimizing parameters of an urban CA model, in particular to a method for optimizing parameters of an urban CA model based on a biogeographical optimization algorithm. Background technique [0002] Since the 1990s, Geo-computation has pushed the study of quantitative geography into a new era. The idea of ​​combining mathematical models and simulation experiments has gradually attracted the attention of scholars. Cellular Automata (CA) As a discrete dynamic model [Document 1-2], it has unique advantages in simulating complex nonlinear problems, and has become an important tool in the research of applied geography and other disciplines. It is used in urban expansion simulation [Document 3-8], image Segmentation [9-10], forest fire spread [11], land use change [12-16] and other fields have been widely used. Urban expansion is a hot area of ​​application of...

AI Technical Summary

Problems solved by technology

These studies show that, compared with the traditional parameter acquisition methods, this method can automatically obtain the optimal model structure and weight parameter combination more efficiently and quickly, and the simulated urban spatial form is more realistic and closer to the actual development of the city. However, such optimization The algorithm may also have problems such as slow convergence speed, low degree of global optimization, and easy to fall into local optimum.

Images