This study focuses on chemical process optimization, focusing on solving the problems of difficulty in processing complex data and difficulty in optimizing the process efficiently in chemical processes. In view of the characteristics of chemical process data, this paper deeply develops an improved principal component analysis algorithm, focusing on the application of kernel principal component analysis (KPCA) in the chemical industry, so as to achieve precise dimension reduction and effective feature extraction of massive and complex chemical process data. After that, the extracted features are used, and systematic analysis is performed to determine the combination of factors that have the greatest impact on the quality of chemical products. By reasonably adjusting operating parameters such as temperature, pressure, and feed rate, the research results are as follows: the product purity can reach up to 94%, and the impurity content is reduced to 62 ppm. In addition, the study points out that the separation link in the chemical process is the bottleneck restricting the overall efficiency. Its principal component contribution rate is relatively evenly distributed, and the separation efficiency is at a low level, which provides a key basis for the optimization and improvement of subsequent chemical processes.