<p>Geomagnetic activity indices, such as the disturbance storm time (DST) index, help identify and analyze periods of geomagnetic storms. Researchers have many suggestions about geomagnetic activity indices. One of the indices of the presence of magnetic storms and their intensity levels is defined using the DST index. This index results from observations made on the ground, and the data is manipulated to obtain a single planetary index. Many studies on the DST index have observed and proposed its prediction methods, including using solar wind parameters. While the physical pathways are complex, recent studies suggest potential links between space weather and tropospheric dynamics. This study, therefore, aims to investigate whether these links can be captured empirically by using the DST index and solar wind parameters as inputs to predict local weather variables in the imaginary equatorial line in Indonesia. In addition, seven areas in this study were observed and crossed by the imaginary line. The BISECANN method is one of the new deep learning methods used in this paper to analyze the correlations and the relationship between the DST index and the influence of weather on the equatorial regions of Indonesia. The dominant factors in the influence of solar wind parameters on weather parameters are density and magnetic fields with a significance level below 0.05. The prediction results are quite good with an R-squared value close to 1, except for Pekanbaru.</p>