Machine learning driving type urban surface heat island intensity mechanism analysis method fused with local climate region framework

By integrating the local climate zone framework and machine learning methods, urban areas are divided into local climate zones. An optimized model is constructed for dual-scale analysis, which solves the problem of accuracy in urban heat island effect analysis and realizes the precision and systematic nature of thermal environment regulation strategies.

CN122332967AActive Publication Date: 2026-07-03SHENZHEN UNIV
4 Cites 0 Cited by

Patent Information

Application Number
CN202610798490.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-06-04
Publication Date
2026-07-03
Estimated Expiration
2046-06-04

AI Technical Summary

Technical Problem

Existing urban heat island effect analysis techniques cannot accurately and systematically reveal the driving mechanisms of surface urban heat island intensity in different local climate zones at the urban agglomeration scale, which limits the precise implementation of thermal environment regulation strategies.

Method used

A machine learning-driven approach that integrates a local climate zone framework is adopted. By dividing local climate zone LCZ type units through multi-source geospatial data, an optimization model is constructed, and a two-scale analysis is conducted to obtain the contribution and interaction effects of urban morphology factors, thereby generating targeted thermal environment regulation strategies.

Benefits of technology

It significantly improves the accuracy and spatial targeting of the analysis of the driving mechanism of urban surface heat island intensity in local climate zones, and provides overall guidance for the whole region and local differentiated control measures.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention provides a machine learning driving type urban surface heat island intensity mechanism analysis method fused with a local climate region framework, and the method comprises the steps: obtaining the geographic space data of an urban agglomeration region, dividing the urban agglomeration region into a plurality of LCZ type units under a classification system of LCZ1-LCZ10 / LCZA-LCZG based on the geographic space data, and obtaining an LCZ classification result; on the basis of the geographic space data and the LCZ classification result, SUHII is determined, and SUHII space distribution data is obtained; extracting two-dimensional and three-dimensional city form factors based on the multi-source geographic space data; based on the SUHII spatial distribution data and the urban form factors, an XGBoost model is trained, and an XGBoost optimization model is obtained; performing dual-scale analysis on the XGBoost optimization model by adopting an SHAP analysis method to obtain a dual-scale analysis result; and generating a thermal environment regulation and control strategy for the whole urban agglomeration and various LCZ units according to the double-scale analysis result.
Need to check novelty before this filing date? Find Prior Art