Feature selection (FS) is vital in hyperspectral image (HSI) classification, it is an NP-hard problem, and Swarm Intelligence and Evolutionary Algorithms (SIEAs) have been proved effective in solving it. However, the high dimensionality of HSIs still leads to the inefficient operation of SIEAs. In addition, many SIEAs exist, but few studies have conducted a comparative analysis of them for HSI FS. Thus, our study has two goals: (1) to propose a new filter–wrapper (F–W) framework that can improve the SIEAs’ performance; and (2) to apply ten SIEAs under the F–W framework (F–W–SIEAs) to optimize the support vector machine (SVM) and compare their performance concerning five aspects, namely the accuracy, the number of selected bands, the convergence rate, and the relative runtime. Based on three HSIs (i.e., Indian Pines, Salinas, and Kennedy Space Center (KSC)), we demonstrate how the proposed framework helps improve these SIEAs’ performances. The five aspects of the ten algorithms are different, but some have similar optimization capacities. On average, the F–W–Genetic Algorithm (F–W–GA) and F–W–Grey Wolf Optimizer (F–W–GWO) have the strongest optimization abilities, while the F–W–GWO requires the least runtime among the ten. The F–W–Marine Predators Algorithm (F–W–MPA) is second only to the two and slightly better than F–W–Differential Evolution (F–W–DE). The F–W–Ant Lion Optimizer (F–W–ALO), F–W–I-Ching Divination Evolutionary Algorithm (F–W–IDEA), and F–W–Whale Optimization Algorithm (F–W–WOA) have the middle optimization abilities, and F–W–IDEA takes the most runtime. Moreover, the F–W–SIEAs outperform other commonly used FS techniques in accuracy overall, especially in complex scenes.

In recent years, with the introduction of the concept of a local climate zone (LCZ), researchers have proved that adding an LCZ to the Weather Research and Forecasting (WRF) Model can improve the simulation effect. However, many existing studies cannot explain whether the improvement of accuracy in the model results is the effect of the refined zone or the effect of urban area correction, so they cannot explain the advantages of LCZ data. Therefore, this paper uses remote sensing images to generate two kinds of land use data sets and introduces them into the Weather Research and Forecasting Model coupled with the building energy model (WRF-BEM). In this paper, the two factors of urban area expansion and fine classification are considered, and three numerical examples are set up to simulate high-temperature weather in August 2019. The research shows that the simulated 2 m temperature of the scheme of correcting only urban area is the closest to the observed data. Although the RMSE in the 2 m temperature simulated by the LCZ scheme is 0.43 °C higher than that of the scheme of correcting only the urban area, it can well reproduce the spatial variation characteristics of 2 m temperature. In addition, different urban morphologies affect the spatial distribution of the surface urban heat islands in Beijing. High surface urban heat island effect zones mainly appear in the compact low-rise, compact mid-rise, and large low-rise types.