With the continuous growth of society, buried pipelines are increasingly widely used in oil and gas transportation, and the safety issues of buried pipelines in geological disasters such as earthquakes have received great attention. The accurate calculation of seismic motion parameters is the basis for seismic safety evaluation of construction sites. In the analysis of seismic performance of structures, the selection of seismic motion directly affects the rationality and accuracy of the structural time history analysis results. There are various seismic intensity parameters that affect the dynamic response of pipeline engineering, and studying the correlation between seismic intensity parameters and structural seismic response indicators is of great practical significance for the seismic design of underground structures. To systematically study the influence of seismic parameters on the dynamic response of pipeline engineering, this paper designs an optimization algorithm for seismic parameters, aiming to improve the resistance of pipeline engineering to seismic and other dynamic responses. Using machine learning (ML) algorithms, the optimal combination of seismic parameters is found by simulating and analyzing the effects of different seismic parameters on the dynamic response of pipeline engineering. The experimental results indicate that this method can provide scientific basis for the design and optimization of pipeline engineering, improve its safety and reliability.